Lenny's Podcast

The rise of taste, human authenticity and judgment in an AI world | Adam Mosseri (Head of IG)

2219 summary words 10 min summary Watch video

Start with the signal

10 min read

Summary

At-a-Glance

  • Verdict: Watch fully
  • Core thesis: As AI commoditizes execution, durable advantage shifts toward small generalist teams, human taste and judgment, carefully bounded agent delegation, and trusted human identity behind increasingly synthetic output.
  • Why it matters: Mosseri provides reusable operating patterns for AI-native product teams, strategy workflows, recommender control, content provenance, and managing experiments at massive scale.
  • Best use: Use it to pressure-test Ken's organizational design, agent orchestration model, human approval boundaries, provenance strategy, and approach to AI-generated content.

Executive Summary

Meta is replacing the traditional roughly 13-person cross-functional product team with six- or seven-person pods: typically four to six increasingly generalist engineers, one “product staff” generalist, and specialists brought in only when the work demands exceptional depth. AI tools let product staff perform previously bespoke design, research, and data-science tasks, reducing coordination overhead and work designed by committee.

Mosseri argues that easier execution raises the value of deciding what to build. Human advantage moves toward taste, vision, strategy, judgment, and the management-like skill of setting goals, constraints, autonomy, and feedback loops for AI agents. AI can clarify strategy, but generic prompts produce predictable answers unless the user supplies technological, personnel, competitive, regulatory, compliance, brand, and budget context—and chooses a model willing to challenge assumptions.

Instagram's historical recommenders did not semantically “understand” users as people commonly assume; they primarily operated through illegible embedding vectors and behavioral correlations. LLMs can now translate areas of embedding space into understandable concepts, enabling Instagram's “Your Algorithm” interface, where users can inspect and modify inferred interests. This creates a path from opaque personalization toward conversational user control.

Mosseri expects synthetic content to benefit Instagram because abundance should increase demand for recognizable people, perspectives, creativity, and authenticity. He opposes filtering content merely because AI made it, favoring quality ranking, account context, and provenance disclosure. The hard limit is detection: as generation improves, platforms may need confidence-qualified labels or even positive verification of camera-captured media rather than assuming all AI content can be reliably detected.

Key Takeaways

  • Claim: AI-native product development favors much smaller core teams composed of broad builders, with exceptional specialists added only when the problem requires them. | Evidence: Meta's prior canonical team was roughly a baker's dozen: two or three engineers each for Android, iOS, and server, plus a PM, designer, data scientist, researcher, and sometimes a generalist. Its new pods usually contain four to six engineers, one product staff member, and any necessary specialist, producing a core of about six or seven. | Implication: Ken should design AI-operations teams around compact, end-to-end ownership while maintaining access to a bench of deep experts rather than embedding every specialty in every workflow. | Caveat: Mosseri does not argue that specialist functions disappear; novel experiences, pricing strategy, and other difficult problems may still require very senior designers, data scientists, or researchers.
  • Claim: Taste and judgment become more valuable as AI lowers the cost of building, because selecting the right problem increasingly dominates mechanical execution. | Evidence: Mosseri says designers may remain unusually valuable because taste is difficult to automate, while AI-generated products already exhibit recognizable default aesthetics associated with tools such as Codex, Claude, Replit, and Lovable. He defines vision as the desired future state and strategy as an opinionated path that a reasonable person could dispute. | Implication: Ken should treat taste, prioritization, and controversial strategic choice as explicit human control-plane capabilities, not as residual tasks left after automation. | Caveat: Taste alone is insufficient; strong contributors also need informed opinions about product strategy, business, and go-to-market rather than remaining confined to visual or interaction design.
  • Claim: AI changes engineering and adjacent roles from direct production toward planning, steering, and review, which will reorder who performs well. | Evidence: Mosseri estimates engineering previously involved roughly 40% to 60% code writing, whereas people in AI labs now spend most of their time planning and reviewing generated code. He cites designers programming, engineers performing analysis, and data scientists creating design proposals; he can again write code despite describing himself as only a mediocre engineer. | Implication: Performance systems should reward decomposition, specification quality, verification, and cross-functional range rather than measuring output primarily through manually produced artifacts. | Caveat: The tools remain highly uneven: impressive at some tasks and remarkably poor at others, so success requires a clear-eyed understanding of present limitations and an informed guess about how quickly those limitations will move.
  • Claim: AI is useful for strategy only when humans supply the complete operating context and deliberately seek adversarial feedback. | Evidence: Mosseri says an unstructured request for strategy tends to return the predictable move competitors would expect. Useful strategy work must include technology, available personnel and motivation, talent attraction, competitors, regulation, compliance, brand identity, goals, budget, and other constraints; he also recommends a model that pushes back rather than behaves like a pleaser. | Implication: Ken's strategic agents should be fed structured context, required to expose assumptions and trade-offs, and evaluated for constructive dissent instead of fluency or agreement. | Caveat: Even with rich context, Mosseri frames AI as a clarifying conversational partner rather than an autonomous owner of strategy.
  • Claim: Managing agents will resemble managing teams: leaders must define success, choose an appropriate autonomy level, and provide feedback without either micromanaging or abandoning direction. | Evidence: Mosseri compares future agent delegation with leadership: excessive prescription stifles useful ideas, while excessive openness allows work to drift. He warns that the ability to outsource a workflow does not mean it should be outsourced and notes that some workflows are win-win while others carry risks greater than their benefits. | Implication: Agent orchestration should encode adjustable autonomy, checkpoints, success criteria, and escalation paths rather than using a universal hands-off delegation model. | Caveat: The interview does not provide a formal risk taxonomy or concrete approval matrix for deciding which workflows should remain human-controlled.
  • Claim: LLMs can convert opaque recommender representations into user-legible concepts, opening a path to direct control over personalization. | Evidence: Instagram's recommenders historically represented interests as vectors and behavioral correlations rather than statements such as “this user likes surfing.” Its “Your Algorithm” feature uses an LLM to describe regions of embedding space with labels such as “deep pour-over coffee snobbery,” then lets users add or remove inferred topics. | Implication: Ken can apply the same pattern to agent memory and routing: translate latent classifications into editable natural-language profiles so operators can inspect and correct the system's assumptions. | Caveat: Current controls are primarily topical; requests involving mood, relationship closeness, format repetition, or other non-topical preferences remain future work.
  • Claim: Synthetic content should be ranked by safety, relevance, quality, and point of view—not rejected based solely on the generation tool—but trustworthy provenance will become essential. | Evidence: Mosseri cites Plastic Dream Sequence and a Paris-based creator producing distinctive dreamscapes with multiple image, video, and music models as examples of AI work with recognizable creative authorship. Conversely, he identifies fake AI personas, such as an AI monk selling bogus supplements, as a new spam vector. | Implication: Systems handling generated media should separate creative quality from provenance risk and attach confidence, account history, and identity context rather than relying on a binary AI label. | Caveat: AI detection may become unreliable as models improve. Mosseri suggests confidence-qualified assessments and says labeling verified camera-captured content may ultimately be more practical than trying to detect every synthetic artifact.

Detailed Brief

Leadership becomes curation and team composition

  • Claims: The strongest product leaders are often not prolific solo visionaries but curators of people, ideas, technologies, and strategies.; The origin of a winning strategy matters less than selecting it, achieving team commitment, and executing it well.; Leadership-team chemistry is a material operating variable: individually strong people may still form a dysfunctional group.
  • Evidence: For an area such as trust and safety, Mosseri evaluates how engineering, product staff, data science, design, and research leads complement one another rather than assessing each role independently.; He says a leadership team with trust and rapport can work through most problems, while almost anything can become an issue when those qualities are absent.; A senior Instagram designer named Nate moved into product staff, illustrating how strong craft specialists can expand into broader product, strategy, business, and go-to-market responsibilities.
  • Caveats: Team chemistry is described as more art than science and therefore difficult to formalize in hiring or organizational metrics.; Meta still needs junior specialist talent to develop into future senior experts; staffing only today's elite specialists would create a long-term capability gap.
  • Implications: AI-era leadership leverage may come less from personally generating answers and more from constructing the environment in which strong options emerge and the best one is selected.; Career architecture should support movement between specialist and generalist tracks without treating broader roles as the only path to advancement.

Platform evolution, experimentation, and costly timing errors

  • Claims: At very large scale, every meaningful product test should be treated as if it will become public before the company knows whether it wants to ship it.; Backlash does not automatically invalidate the strategic direction; it may indicate excessive speed, poor explanation, or conflation with other changes.; Extending an existing product primitive can be less risky aesthetically but strategically wrong when the new behavior requires a different distribution model.
  • Evidence: Instagram's controversial video-oriented feed redesign reached only 4% of iOS users as an early test, but it became conflated with Reels investment, recommendations from unfollowed accounts, and creator concerns about declining reach.; Instagram now discusses controversial experiments in advance around the question “not if it leaks, when it leaks,” including whether to communicate proactively or reactively and what message to use.; The first version of Reels was built inside Stories in 2019. Low story read-through, content overload, and disappearing posts meant most Reels were not seen; Mosseri believes arriving with the standalone mid-2020 version a year earlier could have left TikTok less dominant when pandemic usage exploded.; Facebook Home, an Android-level project involving HTC, carriers, OEMs, and certification, was a spectacular failure but taught Mosseri that executing an idea well enough to disprove market fit can be the right outcome.
  • Caveats: Mosseri's counterfactual about TikTok's potential size if Reels had launched differently is his judgment, not a demonstrated causal result.; Large-scale experimentation creates an unavoidable tension: products cannot safely launch to billions without testing, but tests cannot remain private.
  • Implications: Communication readiness is part of the experiment design itself for high-visibility products, especially pricing, feeds, identity, and other trust-sensitive surfaces.; Teams should distinguish a flawed strategic direction from a valid direction expressed through the wrong primitive, timing, rollout pace, or narrative.

Notable Concepts & Terms

  • Product staff: Meta's evolution of the PM into a generalist who can perform parts of design, data science, and research through AI-assisted tools while owning broader product decisions.
  • Pods: Small, usually six- or seven-person product units designed to reduce coordination costs and committee-driven decisions.
  • Taste: The human ability to recognize what is worth building and distinguish intentional creative work from generic tool defaults.
  • Curator leadership: A leadership model centered on attracting talent, surfacing options, composing teams, and selecting ideas rather than personally originating every answer.
  • Centaur versus reverse centaur: A framing for whether humans direct AI as a tool or become workers mechanically executing decisions supplied by an automated system.
  • Embedding space: A latent map in which similar content appears close together; LLMs can now translate parts of this otherwise illegible space into understandable descriptions.
  • Exploration-based ranking: Showing users uncertain or novel content to discover new interests and give niche creators a chance to find audiences, rather than exploiting only known preferences.
  • Your Algorithm: Instagram's effort to expose inferred interests in natural language and let users modify them, adding agency to recommendation-driven feeds.

Operator Notes / Why Ken Should Care

  • Run a time-boxed pod pilot with one accountable product generalist, four to six builders, and specialists available on demand; compare cycle time, decision latency, and rework against a traditional cross-functional team.
  • Add a mandatory strategy-agent context schema covering objectives, constraints, personnel, incentives, competition, regulation, compliance, brand, budget, and explicit non-goals.
  • Benchmark candidate models for disagreement quality and error detection, not only answer quality; reject configurations that reflexively validate the operator.
  • Define autonomy tiers for agents, including tasks they may complete independently, tasks requiring checkpoints, and decisions that must remain human-owned.
  • Make inferred routing profiles, user preferences, and agent memories inspectable and editable in plain language rather than exposing only opaque scores.
  • Design provenance records with graded confidence, source or capture evidence, account age and history, and identity signals; avoid depending on a permanent binary AI detector.
  • Require a leak-response and stakeholder-communication plan before launching any sensitive pricing, identity, ranking, or interface experiment.
  • Protect the specialist talent pipeline by identifying junior practitioners with credible paths to deep expertise instead of staffing exclusively with senior experts and generalists.

Source/Metadata

  • Title: The rise of taste, human authenticity and judgment in an AI world | Adam Mosseri (Head of IG)
  • Transcript words: 20270
  • Duration seconds: 4109
  • Timestamp note: No timestamps or chapters were present. The supplied extraction contains duplicated and partially repeated passages, so timestamp navigation is unavailable.
Search this video Resume transcript at 0%
Full transcript 12673 words · 90 min read
0:00

SPEAKER_01

I think taste matters a ton. In a world where it's easier to build things, it's more important to make sure that your time is spent figuring out what you should be building in the first place. The people who I think are going to make the most of it are the ones who are clear-eyed about what AI is good at and what it's not good at, and also have an instinct or a nose for what it will be good at and not good at.

0:23

SPEAKER_00

What's something that the Instagram algorithm knows about human behavior that people may not realize?

0:30

SPEAKER_01

I think people assume that there's a much more detailed semantic understanding of everybody's interests and preferences in Instagram than there is.

0:38

SPEAKER_00

Is the rise of AI content a headwind or a tailwind for Instagram versus other platforms?

0:44

SPEAKER_01

I think it's going to be a tailwind, but I think it's going to be a challenge. In a world where there's an abundance of synthetic content, I actually think people are going to seek out creativity and authenticity and people. I don't think we should filter out AI content. I think we should let you know if content is AI content or not.

1:03

SPEAKER_00

That's hard, by the way. Where do you think human brains will continue to be most valuable as AI continues to eat more and more of that product development lifecycle? That's a great question.

1:16

SPEAKER_00

Today, my guest is Adam Mosseri, head of Instagram. Over 3 billion people use Instagram monthly. That's one in every three people alive. It boggles the mind. Prior to Instagram, Adam designed and led the early Facebook newsfeed. He also ran the team that built the Facebook ranking algorithm. And eight years ago, he took over Instagram from its founders, Kevin Systrom and Mike Krieger. He's a designer turned product manager, turned leader of Instagram. Adam is also famous for being the face of all of the controversy and changes that come with evolving Instagram as a product, which we talk about. Before we get into it, don't forget to check out Lenny's product pass.com

1:55

SPEAKER_00

for a free year of the most interesting and well-crafted AI products in the world, available exclusively to Lenny's newsletter subscribers. With that, I bring you Adam Mosseri.

2:09

SPEAKER_00

Adam, thank you so much for being here. Welcome to the podcast.

2:13

SPEAKER_01

Thank you for having me. Excited to be here.

2:15

SPEAKER_00

You've been doing product for a long time. You get to see how a lot of teams operate across Meta within Instagram. What does the canonical product team look like in 2026? What's most different today in how teams operate versus, say, a couple of years ago?

2:33

SPEAKER_01

It's changed a lot this year. [SPEAKER_01] So for the longest time at a big company like ours, the canonical team was something like two or three Android engineers, two or three iOS engineers, two or three server engineers, maybe a generalist, a PM, a designer, a data scientist, a researcher if you were lucky. And maybe that's about it. So on the order of a baker's dozen. And that is a function of wanting to have for anybody who's writing code, someone who can review their code and who's familiar with that code base and having these different functions that are more specialized. I think it's very different at a startup. But this year, it's changing.

3:14

SPEAKER_01

We've adopted what we call pods, which are mini teams, where it's call it four to six engineers who are a bit more generalists. One we call product staff, which is an evolution of the PM. So a PM who can do some of what a designer does and some of what a data scientist does and some of what a researcher does, leveraging the latest tools that we have for them. And then whatever specialist they need. If they're doing something that requires a pricing strategy, you need a senior data scientist. If you're doing something that is really novel from an experience standpoint, you need a very senior product designer. So we try to build a team based on the needs of the work,

3:57

SPEAKER_01

but then end up with a much smaller core, which is more on the order of six or seven usually. And that is a very big shift that's just happening to us this year. But just by virtue of having less people to coordinate, they can often move faster and make better decisions, a little bit less designed by committee. So we talk a lot about AI adjusting and improving productivity, and that's part of it. But I think another part of it is just the small teams. I think they often are just more effective. [SPEAKER_00] This episode is brought to you by our season's presenting sponsor, WorkOS.

4:37

SPEAKER_00

What do OpenAI, Anthropic, Cursor, Vercel, Replit, Sierra, Clay, and hundreds of other winning companies all have in common? They are all powered by WorkOS. If you're building a product for the enterprise, you've felt the pain of integrating single sign-on, SCIM, RBAC, audit logs, and other features required by large companies. WorkOS turns those deal blockers into drop-in APIs with a modern developer platform built specifically for B2B SaaS. Literally every startup that I'm an investor in that starts to expand upmarket ends up working with WorkOS. And that's because they are the best. Whether you are a seed-stage startup trying to land your first enterprise customer,

5:17

SPEAKER_00

or a unicorn expanding globally, WorkOS is the fastest path to becoming enterprise-ready and unblocking growth. It's essentially Stripe for enterprise features. Visit WorkOS.com to get started or just hit up their Slack where they have actual engineers waiting to answer your questions. [SPEAKER_01] Go to WorkOS.com to make your app enterprise-ready today. I love this.

5:43

SPEAKER_01

[SPEAKER_00] So on this team of six to seven, what's the makeup again?

5:47

SPEAKER_00

And which role are you finding you have less of if you're going from the kind of—if you're going 50% size? [SPEAKER_01] You just have less specialists, right? [SPEAKER_01] So you might not have any.

5:59

SPEAKER_01

You might be four engineers and a product staff. And there's no data scientist. There's no designer. There's no researcher. There's no content designer. [SPEAKER_00] The product staff is the generalist that sort of supports all of those things. I mean, what's clearly happening is all of the functions are starting to bleed into each other.

6:10

SPEAKER_00

So on this team of six to seven, what's the makeup again?

6:14

SPEAKER_01

[SPEAKER_00] And which role are you finding you have less of if you're going from the kind of the, if you're going 50% size? You just have less specialists, right? So you might not have any. You might be four engineers and a product staff. And there's no data scientist. There's no designer. There's no researcher. There's no content designer. [SPEAKER_00] The product staff is the generalist that supports all of those things. What's clearly happening is all of the functions are starting to bleed into each other. And the whole industry is wrestling with what that means. A lot of what a data scientist does at a big company, for instance, is relatively mechanical.

7:23

SPEAKER_01

So there's stuff that they do that is really more art and science and the stuff that's really more just pulling data, data management.

7:32

SPEAKER_00

[SPEAKER_01] So some of the tools that we're building internally to understand, for instance, a traditional data science question would be a waterfall. [SPEAKER_01] So if you wanted to look at people creating reels, you would look at all the steps and then how people fall off on each step and try to figure out where there might be opportunities to improve things. [SPEAKER_01] That kind of basic waterfall analysis is much easier now to use some of our internal tools to just pull automatically as opposed to having to have a data scientist do much bespoke work for that. [SPEAKER_01] So our product staff might be able to do that now.

7:41

SPEAKER_00

[SPEAKER_01] And they couldn't do that a year ago. [SPEAKER_01] So you just end up with this generalist.

7:54

SPEAKER_01

These people have more generalist shapes. And then when you need it, when you really need it, you have a more senior, ideally, or just more creative specialist.

7:57

SPEAKER_00

[SPEAKER_01] So a phenomenal product designer or a genius data scientist or researcher.

7:59

SPEAKER_01

[SPEAKER_00] This is so interesting. [SPEAKER_00] It's exactly what I just heard. [SPEAKER_00] I had Fiona Fong, the head of engineering for Cloud Code and Cowork on the podcast. [SPEAKER_00] She's Forest Journey's manager.

8:05

SPEAKER_00

And she described the people she hires now are, one, builders with great taste that can take an idea from end to end and people with deep expertise in a very specific domain.

8:05

SPEAKER_01

[SPEAKER_00] The taste thing matters a lot. I really agree with that. Boris used to work at Instagram. [SPEAKER_00] Oh, that's right. Yeah, he was a senior IC at Instagram for a while. I love seeing him. He's all over Threads now. He's the face of Clark Cove. [SPEAKER_00] He's killing it. He's a celebrity now. He really is. Yeah, for sure. In a world that is on fire right now.

8:58

SPEAKER_00

[SPEAKER_01] No, I think taste matters a ton. [SPEAKER_01] So in a world where it's easier to build things, it's more important to make sure that your time is spent figuring out what you should be building in the first place. [SPEAKER_01] Actually, so a lot of designers right now are very anxious about their roles. [SPEAKER_01] You've got these other generalists doing design.

9:10

SPEAKER_01

You've got engineers doing design, product staff doing design. But I'm actually pretty long on design or designers because they tend to have taste. And I think that is something that is much more difficult to imagine being automated away.

9:16

SPEAKER_00

[SPEAKER_01] And so there's other challenges with design sometimes, but I'm pretty long right now on designers. I've always felt that too, as it is so easy to build. And all the work that AI produces is so, you can tell this was cloud design. I know what you did here. This is Codex. [SPEAKER_01] Well, they all have their vibe, right? [SPEAKER_01] Yeah, exactly. [SPEAKER_01] Your vibe code, your apps, we call it vibe code, and you're like, oh, that's a Codex app. Oh, that's a cloud app. Right.

9:40

SPEAKER_01

[SPEAKER_00] And that's replete. [SPEAKER_00] That's lovable. [SPEAKER_00] You can predict these things. [SPEAKER_00] I've always thought that design should be thriving right now. [SPEAKER_00] For some reason, it hasn't yet. [SPEAKER_00] If you look at jobs for designers, they're kind of flatlining. [SPEAKER_00] I feel the missing piece is the PME piece of deeply understanding the business and what will grow it and what successful, all that stuff. [SPEAKER_00] The business side of it versus the taste side of it. Yeah. I think you're going to see, we have a senior designer at Instagram called Nate who just transferred into product staff.

10:52

SPEAKER_01

So I think some of what you'll see is it will be harder to talk about design roles and who's a good designer because they're not going to just stay in traditional design roles. If you're an amazing designer, you probably have strong opinions outside of just the interaction and visual design.

11:05

SPEAKER_00

[SPEAKER_01] You probably have strong opinions on product strategy, even on the business, on the go-to-market. [SPEAKER_01] And so I actually think some of our strongest product staff are going to be converts from design and from data science who are just looking to expand their reach. [SPEAKER_01] And they were influential across functional boundaries before, but this world where those functional boundaries are just wildly blurred just allows them to jump in. [SPEAKER_01] And so sure, they'll be technically a generalist on paper, but they clearly have a uniquely strong ability in one type of craft.

11:12

SPEAKER_00

[SPEAKER_01] But they've got the ability and strong opinions to make informed decisions across other parts or other crafts.

11:17

SPEAKER_01

And so I don't know that all the strongest designers I have will all be in design. They probably will be the majority, but I can imagine a bunch of really strong ones moving roles. But I should also check my own bias here, because I started as a designer at Facebook way back when, and I switched roles. [SPEAKER_00] No, designers are great. [SPEAKER_00] I'm a big fan. [SPEAKER_00] So this is really interesting. [SPEAKER_00] There's always been this GM model where different types of functions can become GMs. [SPEAKER_00] It's this product staff role feels like a similar situation where different functions can become product staff. Yeah, yeah.

11:53

SPEAKER_01

And that was true of PM before, but it's so much more true now.

11:55

SPEAKER_00

[SPEAKER_01] And it'll, I mean, in some ways it's probably the age of the generalist, but I still think there's going to be a real important role for these really amazing specialists who are just all about going. [SPEAKER_01] I wish I was like that. [SPEAKER_01] I always romanticized the phenomenal machine learning engineer or AI researcher or shoemaker. [SPEAKER_01] I think that's the coolest thing in the world. [SPEAKER_01] But it's never been my shape. [SPEAKER_01] I've always been I've never been great at anything. [SPEAKER_01] I've always just had range. [SPEAKER_01] Yeah, yeah. [SPEAKER_01] And that was true of PM before, but it's just so much more true now.

12:26

SPEAKER_01

And it'll, in some ways it's probably the age of the generalist, but I still think there's going to be a real important role for these really amazing specialists who are just, they're all about going. I wish I was like that. I always had this, I romanticized the phenomenal machine learning engineer or AI researcher or shoemaker. I think that's the coolest thing in the world. But it's never been my shape. I've always been, I've never been great at anything. I've always just had range. That's always been my strength. [SPEAKER_00] Same. [SPEAKER_00] Okay. [SPEAKER_00] So this idea of product staff. [SPEAKER_00] So the idea is on these new pods.

13:22

SPEAKER_01

[SPEAKER_00] So this is a new thing you guys are doing. [SPEAKER_00] So there's these pod teams, product staff, engineers, and maybe one specialist that's going deep on say pricing algorithm or something like that. [SPEAKER_00] So what this tells me is there's these adjacent roles that are maybe more in trouble over the years. [SPEAKER_00] Data science, for example, user research, for example, you talked about designers being anxious. [SPEAKER_00] Is there anything there of just like, oh, these maybe folks in these groups should think about shifting to other roles. I mean, there's anxiety everywhere.

13:51

SPEAKER_00

[SPEAKER_01] I've talked to a lot of people at a lot of other companies and it just seems like there is a lot of concern right now about competition, about job displacement, about unintended or unforeseen consequences of all this technology and all this moving so quickly. [SPEAKER_01] So that's definitely happening. [SPEAKER_01] I think that you will see the functional lines continue to blur, but I still think there will be room for functions. [SPEAKER_01] They'll just be shaped differently.

13:56

SPEAKER_01

They won't all be senior, necessarily, but they'll all be either senior or on their way to being senior. You can't just have a bunch of super senior data scientists and no new ones.

14:00

SPEAKER_00

[SPEAKER_01] Because then who's going to be the new super senior data scientists in the future? [SPEAKER_01] So you need to hire and mentor and grow talent. [SPEAKER_01] Maybe the team is smaller overall. [SPEAKER_01] And then those who aren't on their way to being super senior move into more of a generalist role. [SPEAKER_01] I think that's a reasonable soft landing.

14:17

SPEAKER_01

But I do think you're going to want to make sure you're investing not only in today's senior talent for each specific function, but in tomorrow's. Otherwise, I think you're going to regret it in a couple of years. That said, who knows what the world looks like in a couple of years. So my big thing is generally don't be overly confident in whatever your predictions are, because there's just too much flux right now. [SPEAKER_00] Yeah.

14:34

SPEAKER_01

[SPEAKER_00] Ben and Dick DeVidence was on the podcast recently and said the same thing. [SPEAKER_00] We don't know anything about what's going on. [SPEAKER_00] Yeah. I like him a lot. I'll make sure I listen to the pod. [SPEAKER_00] Yeah. [SPEAKER_00] So you talked about taste. [SPEAKER_00] This makes me think about, so you're interviewing a lot of people, hiring a lot of people. [SPEAKER_00] What are some traits that you're trending up in things that you look for more and more now in this world? [SPEAKER_00] And what are some traits that are trending down and maybe less important to you? I mean, there are some things that are the same, right?

15:26

SPEAKER_01

For the longest time, almost no matter what the function, I always look for three things. Do you have grit? You know, you're really going to, you've got some drive, some fire in your belly. Are you a quick learner? And are you reasonably, ideally very self-aware so that you can actually take feedback and know what you're good at and what you're not good at? Because if you've got fire in your belly, you learn quickly and you're self-aware, you can get good at anything eventually. But if any of those things are missing, it was usually an issue.

16:01

SPEAKER_01

So that's the baseline right now for hiring, but for people who are going to be more successful over these next five or ten years as things change so significantly, I think two things that I'm continuing to encourage myself to do are stay curious and put yourself out there. You have to try things, right? To that point before that no one really knows what's going on. You just have to be willing to try things. It's almost, I don't know if you speak another language. Russian. Yeah. Yeah.

16:46

SPEAKER_01

So when you learn another language, I think one of the most important things, one of the best predictors—this is my guess. I don't have any research on this—about are you going to get good at speaking is, are you willing to sound like an idiot?

16:48

SPEAKER_00

[SPEAKER_01] Are you willing to say it and be corrected and not be offended and then just get better and better? [SPEAKER_01] You just put yourself out there and with all of these new tools and models and technologies, I think you just have to be willing to try stuff. [SPEAKER_01] So if you're curious and you try stuff, you'll learn, you'll adapt, but if you're not curious or you're not willing to make mistakes or try things, I think you're in a ton of trouble or these are going to be a really difficult time.

17:07

SPEAKER_00

[SPEAKER_01] So those I think are premiums, not just for hiring at a company like Meta or a team like Instagram, but I just think across the industry and multiple industries over the next ten to twenty years. [SPEAKER_01] Is there something that maybe we're looking for less of for some of these functions?

17:12

SPEAKER_01

I think that there's some that are still going to be very large teams. And so you need people who are really good at managing large organizations. Large organizational leadership is its own craft and skill. It's actually different than management. But I do think there'll be less of those roles. I think we'll have more smaller teams and there'll be less people who manage thousands of people. So that job will not go away, but that will be less of what I'm looking for in hires. Because I'm going to have less roles like that.

17:33

SPEAKER_01

[SPEAKER_00] Something I'm hearing from a few folks is AI is almost resetting people's impact and success in terms of some people that were maybe low performers, pre-AI. [SPEAKER_00] Can now do things they were bad at or AI now allows them to do. [SPEAKER_00] And now they're thriving, building all these things, helping other people. [SPEAKER_00] Do you see that at all? [SPEAKER_00] Just like AI is lifting other people up, maybe lowering some people down. I think we'll have more smaller teams and there'll be less people who manage thousands of people. And so that's not that that job will go away, but that will be less of what I'm looking for in hires.

18:12

SPEAKER_01

Cause I'm gonna have less roles like that. [SPEAKER_00] Something I'm hearing from a few folks is AI is almost resetting people's impact and success in terms of some people that were maybe low performers, pre AI. [SPEAKER_00] Can now do things they were bad at or AI now allows them to do. [SPEAKER_00] And now they're thriving, building all these things, helping other people. [SPEAKER_00] Do you see that at all? [SPEAKER_00] Just like AI is lifting other people up, maybe lowering some people down. Yeah. I mean, the job is just different. I mean, take engineering. Engineering used to be maybe not majority, but a large percentage, 40, 50, 60% writing code.

18:58

SPEAKER_01

It's not now, especially if you talk to anybody in these labs, they're spending most of their time planning and reviewing code. That is a very different job. You might hate that. And you might have loved just writing code, or you might love that. And you might not have been that fast at writing code. So who succeeds is a function of whose strengths are aligned with the tools needs and the business needs. And so this is definitely happening. Another thing is you've had people who had good ideas about how to contribute to other functions, but didn't have the mechanical or technical skills to do so. [SPEAKER_01] And AI reduces the boundary to do that.

19:37

SPEAKER_00

[SPEAKER_01] And then all of a sudden they can. For me, it's funny. [SPEAKER_01] Cause when I got hired at Facebook, all the designers had to be able to program.

19:52

SPEAKER_01

That was our requirement. I went through a technical loop. We gave up on that cause it was too hard to hire people. But I now get to program again for the first time in maybe 10 years.

19:57

SPEAKER_00

[SPEAKER_01] And I am not a good engineer. [SPEAKER_01] I'm a mediocre engineer on a good day. [SPEAKER_01] But now I can write code responsibly, which is just an amazing thing.

20:02

SPEAKER_01

You're seeing this across all sorts of levels and seniority and functions—designers who are programming, engineers who are pulling data and doing strong analyses, data scientists who are putting together proposals for designs. You know, the tools aren't all great, by the way. I think too often we have this really polarized binary outlook on the state of AI. Like, are you AI pilled or are you anti AI? It's like people aren't binary. I said that to the team yesterday. And the state of the tools isn't binary either. You know, they're amazing at some things and remarkably bad at others.

20:58

SPEAKER_01

And the people who I think are gonna make the most of it are the ones who are clear eyed about what AI is good at and what it's not good at. And also have an instinct or a nose for what it will be good at and not good at, not next month or in a couple months from now. [SPEAKER_00] You mentioned that AI writes all our code now.

21:04

SPEAKER_00

Someone tweeted this idea that's stuck with me for months now of just like, remember, we used to be able to just write code for free. [SPEAKER_01] I think you'll still be able to write code for free. [SPEAKER_01] Just do it with a smaller model, but yes. [SPEAKER_01] Yeah. Like, I guess that's true. Like there are models that are close to free, but it's crazy. Now it's just like. [SPEAKER_01] But just think about the cost, think about what you pay for a model now and how much level of intelligence you're getting from that model, and then at that same price point a year ago, what were you getting? [SPEAKER_01] At some point the incremental value just won't matter.

21:25

SPEAKER_01

You know, we're getting there, I think with small projects and programming. I think the models will matter even beyond this week you've got Grok and obviously Claude from Anthropic, but I spent a lot of time with that this week.

21:27

SPEAKER_00

[SPEAKER_01] For the first time I'm like, oh, I'm just talking to a much more technical, much smarter engineer than I am. [SPEAKER_01] The next version, a year out of that model, do I need to pay for frontier tokens for whatever Anthropic model 6.0 is. [SPEAKER_01] Or is Claude just fine for all of my side projects? [SPEAKER_01] Probably just fine. [SPEAKER_01] Probably pretty cheap by then too. Yeah. When Kevin Weil was on the podcast, when he was CPO at OpenAI, he famously said, this is the worst the model will ever be.

21:38

SPEAKER_01

[SPEAKER_00] Yeah. [SPEAKER_00] It's still hard to comprehend that. [SPEAKER_00] That's only going to get better. [SPEAKER_00] So on this point of tokens, spend ROI and things like that, Meadow was famous for this leaderboard of token spend. It's a terrible idea. No leaderboards for tokens.

22:29

SPEAKER_01

[SPEAKER_00] Okay.

23:02

SPEAKER_01

[SPEAKER_00] Talk about that. [SPEAKER_00] And how do you think about budgets for engineers and product teams at this point? [SPEAKER_00] Just spend as much as you want? [SPEAKER_00] Is there a cap we have?

23:23

SPEAKER_00

Is there anything you've figured out that works well? [SPEAKER_01] Right now we've managed to get the costs reined in a little bit by shutting down the silly things that we were doing. [SPEAKER_01] And so it's not that hard to build a token incinerator and that doesn't create a lot of value. [SPEAKER_01] And as soon as you actually look at the dollars in and value out, you might just be like, oh, that's just a bad idea. [SPEAKER_01] And so right now we don't have token limits for our engineers. [SPEAKER_01] I think for anybody really. [SPEAKER_01] I think for the headcount across my teams.

24:04

SPEAKER_01

[SPEAKER_00] Is there a cap we have? [SPEAKER_00] Is there anything you've figured out that works well? Right now we've managed to get the costs reined in a little bit by shutting down the silly things that we were doing. And so it's not that hard to build a token incinerator and that doesn't create a lot of value. And as soon as you actually look at the dollars in and value out, you might just be like, oh, that's just a bad idea. And so right now we don't have token limits for our engineers. I think for anybody really. I think for a headcount across my teams.

24:59

SPEAKER_01

There I think costs will go up because we'll just be using more tokens not because prices will necessarily go up but then I think prices will come down because all of these frontier models are gonna be in a bit of a pricing war so we'll see. I think it'll be a bit of a roller coaster.

25:05

SPEAKER_01

[SPEAKER_00] So coming back to this idea that as you said we've evolved from we used to write all our code to now we're approaching all code will be written by AI and it feels like now the transition is it's not just written by AI but it's one-shotted by AI like coding now is steering AI and it's how often you have to correct it is coding now and then there's the software development life cycle slowly being eaten by AI. It'll start helping us come up with ideas I imagine more and more. The question I like to ask people is where do you think human brains will continue to be most valuable as AI continues to eat more and more of that product development life cycle?

25:13

SPEAKER_01

Taste. Like we talked about judgment particularly around strategy right like you might get feedback from an AI on the strategy but you're not asking an AI to come up with a strategy anytime soon or if you are then it's within the context of bounds you set so here's my goal here's my vision here are my constraints here's my job here's my budget. I think that you know it looks more like management right like you are trying to define what success looks like decide how prescriptive you want to be about the path to success and then giving feedback along the way and that is its own craft. You know when you it'll be interesting to see how, you know, some of the same dynamics come up like I believe that if you are too prescriptive as a leader with a team you end up stifling good ideas but if you're too open-ended sometimes teams just waste time going in the wrong direction and so that level of autonomy you give a team like maybe that applies to main agents in the future particularly when we're talking not just about building something but deciding what you build in the first place. But I think of vision as an articulation of the world or the state of the product you want to get to and I think of strategy as an opinionated path to achieve that vision. If strategy can't be the best or be amazing it has to be controversial. You have to be or just a reasonable person should be able to disagree with it because otherwise you're probably just trying to compete on raw execution and I think that both vision and strategy I think are going to be where our brains are spreading a lot of our more and more of our cycles and I think less on execution.

25:17

SPEAKER_01

[SPEAKER_00] Something I have always thought is AI should be incredibly good at strategy because you would think here's the market here's all the information on the market our competitors our metrics our numbers our growth all these things help me figure out how to win. You think AI knowing all that would be really good at this.

25:23

SPEAKER_01

I think it could be. I have found it's not unless you steer it pretty aggressively and I don't mean towards an answer I mean based on the constraints. It turns out when you're trying to come up with a strategy there's a lot of things to consider right. You need to consider the state of the technology the personnel on the team and what's motivating them and what you can get. You know sometimes it's coming up with an idea that is on the bubble, you know it's going to actually attract some of the best talent and so that kind of like that kind of the push then goes to the idea obviously the competitive landscape the regulatory landscape for companies as large as ours and the compliance landscape the identity and reason to exist for the brand. You know you have to consider all of these things. I think if you ask an AI just for a strategy lazily you're not going to get something right. You're going to get something pretty predictable that pop up with the competition would expect you to do. I think if you want a really more effective one you need to think long and hard about what are all the different inputs that need to be considered make sure you steer the AI in a way that it's considering those as well and it needs to be a conversation back and forth but I think if you're willing to put in the work and the time it can definitely be helpful and definitely be clarifying. Particularly if you tell it to be critical. Different models have very different vibes though on how willing they are to be pushed back so I recommend picking one that likes pushing back.

25:29

SPEAKER_01

[SPEAKER_00] Yeah Mythos has gotten really good at being like I can't do this let's move on. There's all these what has always been a little bit of a jerk in a way that I actually appreciate. I appreciate it. I really do because I don't want one that's just like oh you're so right I'm so sorry I said that. It's like hold on I want I want you know I want the real intelligence I don't want no pleaser. This point you made about people being excited about the strategy such an interesting

25:42

SPEAKER_01

helpful and definitely be clarifying, particularly if you tell it to be critical. Different models have very different vibes though on how willing they are to be pushed back, so I recommend picking one that [SPEAKER_00] likes pushing back. Yeah, Mythos has gotten really good at being like, I can't do this, let's move on.

25:55

SPEAKER_00

[SPEAKER_01] There's all these—what has always been a little bit of a jerk in a way that I actually appreciate. I appreciate it, I really do, because I don't want one that's just like, "Oh, you're so right, I'm so sorry I said that." It's like, hold on, I want the real intelligence. I don't want a pleaser.

26:02

SPEAKER_00

This point you made about people being excited about the strategy is such an interesting one. There's this idea that I read—I think Corey Doctor wrote this—there's this concept of a centaur and a reverse centaur. So a centaur is a human body horse—this is going somewhere, I promise—human upper part, horse lower part. Horse body, yeah. Horse body where the human is in charge, and that's what we prefer. We want to be in charge. Reverse centaur, which is what we want to avoid with AI, is where the AI is controlling us and we're just doing its bidding. It's a horse head on a human body. [SPEAKER_01] Yeah, exactly. It's terrifying.

26:13

SPEAKER_01

[SPEAKER_00] So in a sense, Uber drivers and DoorDash people kind of—this is their life, which is not great. And this is the danger thing for a lot of people: if it's giving us the strategy and telling us, "Here's what we're doing"—no one's going to want to do that. So that's a really interesting counterpoint to the idea that we don't want AI to be telling us the strategy.

26:19

SPEAKER_01

Yeah, no. I think there's a lot of things to be careful about right now, and I would certainly not just assume that because you might be able to outsource some workflow to AI that you should. There are certain ones where I think it's really just a win-win. There are certain ones where I think the risk outweighs the benefits. This episode is brought to you by Mercury.

26:24

SPEAKER_01

[SPEAKER_00] Radically different banking, loved by over 300,000 entrepreneurs, and now with Command. I've been a customer of Mercury's for over six years. I have never once thought about leaving. Mercury is basically what happens when banking is built by product people, not by bankers. They make it so easy—dare I say fun—to send invoices, move money around, set up virtual cards for folks on my team. Does your bank have an API, a terminal native CLI, or an AI-ready MCP server? I don't think so. And just recently, they launched Command, a conversational interface built directly into Mercury, which acts as your financial operator. I've been using Command to transfer money around, to figure out what categories I've been spending the most money in, analyze my cash flows, and just today I used it to find out how much I've made from a specific sponsor over the past year. I just asked, "How much have I made from X over the past year?" Ten seconds later, I have an answer. It is so cool. Visit mercury.com to learn more and apply online in minutes. Mercury is a fintech company, not an FDIC-insured bank. Banking services provided through Choice Financial Group and Column NA members FDIC.

26:29

SPEAKER_01

[SPEAKER_00] Okay, going back to product leadership and things you've learned along your journey. We were chatting ahead of this about things you've learned, and one thing that you said about some of the best product leaders you've worked with is that they're less visionary and more curators. I'd love to hear more along these lines.

26:33

SPEAKER_01

Yeah, I mean, you do sometimes find amazing product leaders who are idea machines—just prolific idea machines. But I do think a lot of the best ones have taste and have something about them that really makes strong talent want to work with them, but end up being curators. Curators of people, curators of ideas, curators of technologies, curators of strategies. Because I don't really care if I'm hiring a strong lead for an area whether the strategy comes from them or comes from somebody else. I just care that there is an amazing strategy and everyone is bought into it and that we're executing against that strategy well. So I think that some of the best product leaders, yes, have ideas—it's hard to be a great curator if you don't have some of your own ideas—but they embrace the reality that they can't come up with everything themselves. So they need to create an environment in which great ideas bubble up and are chosen or decided upon. So it's not just about curating ideas, but it's sometimes about curating teams and people.

26:38

SPEAKER_01

[SPEAKER_00] I love that. I so agree. I feel like everyone's always joining a team and they just want to do vision and strategy, just not actually hands-on work. And no, AI's coming in here to do the strategy. Yeah, exactly. And I love this point that there's so much power and value, and people underestimate just the need for a really good curator of the team's ideas.

26:51

SPEAKER_01

Yeah, sometimes it's not just who's good or what idea is good. It's also what is going to work given the broader context. So, for instance, on team building, a huge thing that I'm always considering is not just whether this person is a really strong candidate for this role, but how does this person fit into their leadership team. So if I'm in an area like trust and safety, I have an engineering lead, I have a product staff lead, I have a data science lead, I have a design lead, I have a research lead. I need to make sure that those five complement each other. I need to make sure that's about what skills each one has, what weaknesses each one might have. I also need to make sure that they—this is more art than science—have a good vibe. You need trust and rapport. A leadership team with strong trust and rapport can work through most anything. A leadership team without trust or rapport—anything can become an issue. So that chemistry bit is much more art than science, but it matters. So I think some of the best leaders, and product leaders specifically, also either do that instinctively or consciously, but they have a nose for building teams that are going to have good energy and good collaboration.

26:57

SPEAKER_01

[SPEAKER_00] Yeah, well, this flip is also true, right? I've had many times in my career where I've had two people who I think are amazing, and I even adore them and love them, and they just can't get along. You're like, "This isn't a competency issue. This is just a personality issue," and you just have to call it and split them sometimes. [SPEAKER_00] I want to transition to talk about Instagram—the product, the platform, things you guys have learned there. Let me start with this question: What's something that the Instagram algorithm knows about human behavior that people may not realize?

27:08

SPEAKER_01

One of the most common misconceptions is actually in the opposite direction. I think people assume that there's much more detailed semantic understanding of everybody's interests and preferences in the algorithm than there is. Most of what's really driven the progress in the world of recommenders over the last five, ten years have been these large embedding models and other techniques that basically produce artifacts that cannot be read by people. They're not legible. They're like giant vectors. It's like, sure, I can show you the vector, but it's just going to be a bunch of numbers.

27:13

SPEAKER_01

[SPEAKER_00] the platform things you guys have learned there let me start with this question what's something that the Instagram algorithm knows about human behavior that people may not realize?

27:19

SPEAKER_01

One of the most common misconceptions is actually in the opposite direction. I think people assume that there's a much more detailed semantic understanding of everybody's interests and preferences in the algorithm than there is. Most of what's really driven the progress in the world of recommenders over the last five, ten years have been these large embedding models and these other techniques that basically produce artifacts that cannot be read by people. They're not legible. They're like giant vectors. It's like, sure, I can show you the vector, but it's just going to be a bunch of numbers in like a seven-dimensional space. And so when we talk about does the algorithm know something, usually we think in these more semantic terms. It knows I like surfing. And it's like, it doesn't. It just has this big number that happens to correlate with surfing. That said, I think that is starting to change. I think one of the things that LLMs are enabling is they can describe in words, English or whatever language you prefer, what some of those previously illegible artifacts are, at least approximate to, if not directly. So this is the thing I've been really posting about this week—this thing called "Your Algorithm." Basically, the idea is we take a look at all of the stuff that you've interacted with, and all of that is in an embedding space. You can think of embedding space as a map. You can map a bunch of videos into the same map, and videos that are close are similar. And now we can just have an LLM just describe that part of the map, and it can be like, "Oh, that is deep pour-over coffee snobbery," and that's kind of amazing.

27:24

SPEAKER_01

[SPEAKER_00] That's so cool—you can ask the LLM to look at these numbers and extrapolate here's the topic that you're interested in?

27:29

SPEAKER_01

Yeah, or look at the videos. And so both—you can also embed concepts into that same space. And embeddings are really the underlying technology underneath LLMs, right? That's how the whole thing works. And so what we do now is we let you see your algorithm. You can see what topics we think you're interested in, and you can adjust it. You can add and remove things. But the idea here is giving people some agency back in a world where these social media apps are getting taken over by recommendations. But we can't do a lot of other things yet, which we will be able to do. You could ask for things that aren't topical. I want more fun content. I want to see my friends more. I don't want to see my high school acquaintances' photos. We can come up with it. I don't want to see seven photos in a row, but I'm happy to see six photos, or whatever you can come up with. So we have a lot of work to do, and I'm excited about that. But I think a misconception historically is—until recently we don't really know as much about you as you think. We're just like, "You liked these photos. These people also like those same photos, and they like these other photos, so you might like those other photos." That's how it worked. Now only are we actually getting as sophisticated as I think people have assumed we've been for many years.

27:35

SPEAKER_01

[SPEAKER_00] That is really interesting. One that comes to mind is this transition everyone eventually goes through to this algorithmic, broad, global feed. Everyone always feels like people think, "I just want to see chronologically everyone I know and follow, and that's going to be my favorite feed," and it continues to be proven wrong. You actually engage a lot more when it's this algorithmic feed of things we think you will love.

27:42

SPEAKER_00

[SPEAKER_01] Yeah, it's tough because I get it. I posted this week about agency and I just got destroyed in the comments, which is just part of the job. I get it. But there are a couple issues with a chronological feed. One is, and some of this is the tension between an individual's interests and what works when you scale it up. If you do a pure chronological feed, the incentive for everybody is to just post as much as possible because it will always be at the top of everyone who follows you as soon as you post. So what ends up happening is the feed gets overwhelmed with professional content, usually large company content and publishers, because the New York Times can pump out fifty things a day. Your best friend won't. You might get one thing a week from them, and so your feed just gets taken over. Part of it is the incentives that emerge because when you design these systems, it's almost like designing a city. You need to think about, okay, here's how the mechanics work. What are the incentives that arise? How are people going to act within those incentives? And then what happens? And the other thing is, sometimes the most interesting thing was just not the most recent thing. Recency is an important input into relevance, but it's not the only one. My sister got engaged last night, and she's in Germany. But if she did—she's married, she got married last year, that's why it was tough in mind. But if she got engaged, and I missed it because she lives in Europe with different time differences, do I really want to see a picture of my brother's po' boy sandwich, or do I want to see my sister's things first? So it's tough. It's tough. I'd love to figure out a way to find the right balance. I want to give people agency over the experience, but I think it needs to be in a way that creates a system that makes sense, not just for us as a business, which matters—I'm not pretending that's not an issue—but also for the overall community. We've done chronological by default, and where you can make it default, you see not only does usage go down, overall sentiment goes down. The individual who made that choice might be happy at the moment, but when you just get pummeled with stuff you're less interested in over the course of months, we run surveys at massive scales, and we just see people start to become less and less satisfied with Instagram.

27:48

SPEAKER_01

[SPEAKER_00] Kind of along these lines, everybody asks you about this these days—AI and content and how that all impacts everything. I want to ask you something I haven't seen someone ask you: Is the rise of AI content a headwind or a tailwind for Instagram versus other platforms? Do you think this helps or hurts you guys?

27:54

SPEAKER_01

I think it's going to be a tailwind, but I think it's going to be a challenge. And not just because it's more content—obviously, we're an attention-driven business, an advertising business. More content means potentially more attention. That's not for free though. I don't think we're very good at ranking AI content yet. Great AI content versus crap AI content. You should just see the stuff you're interested in and not any of the stuff you're not interested in. But I do think that in a world where, or for years now, and I've said this many times, power is shifting from institutions to individuals across industries.

27:59

SPEAKER_01

Think this helps or hurts you guys. I think it's going to be a tailwind, but I think it's going to be a challenge. It's not just because it's more content. Obviously we're an attention business, a driven business. We're an advertising business. More content means potentially more attention. That's not for free, though. I don't think we're very good at ranking AI content yet. This great AI content, this crap AI content. You should just see the stuff you're interested in and not any of the stuff you're not interested in. But I do think that in a world where, for years now, and I've said this many times, power is shifting from institutions to individuals across industries. The easiest example is sports, where players are more relevant than teams now, and that was not the case when I was a kid. In that world, you're going to—I think it behooves us to invest in individuals and to invest in, specifically for Instagram and creators, and I mean creators broadly. I don't just mean influencers who are promoting branded content and making native-only videos. I mean anybody who's using platforms like Instagram to help do what they do, right? It could be you could be a journalist. You could be an artist. You could be selling scarves you sew. But you're out there as yourself, creating and sharing content that helps you achieve whatever it is you're trying to do.

28:05

SPEAKER_00

[SPEAKER_01] So we've been leaning in that direction for many years now. That's been our, one of our two or three most important audiences for as long as I've been on Instagram. In a world where there's an abundance of synthetic content, I actually think people are going to seek out creativity and authenticity and people more, not less. And I think that will help us. That doesn't mean that we won't have AI content on our platform. There's going to be bad and good AI content, and we're going to try and handle that the way we normally handle content. So unsafe goes away. Interesting versus not interesting is based on ranking and personalization. But I think people are going to really seek out other points of view because Instagram was never just about the content. It was always about, to a certain degree, the person behind the content, the point of view, the reason they're sharing it, their perspective. And I think that's going to become more important, not less. And I think, given that we are not the best at a lot of things, but we are the largest creator platform, you know, if you look at how we define creators and how many creators use us versus other platforms, I think it'll be a tailwind for us because I think people are going to seek out people.

28:11

SPEAKER_00

And this connects to your earlier point that companies, say New York Times, can pump out a bunch of AI content versus a creator. And so you're saying you want to protect against that to allow individuals to continue to perform well in spite of all the AI content?

28:16

SPEAKER_00

[SPEAKER_01] If you just love AI content, great. You should be able to have a feed that's just AI town. And if you don't, then you shouldn't have it in your feed. To me, I don't think we should—I mean, I understand why people are right. There's a paradigm shift and revolution that we're sitting in. But I don't think we should judge content based on the tool that made it. I think we should judge it based on the content, the point of view, the person behind the content. I don't think we should filter out AI content. I think we should let you know if content is AI content or not. I think we should let you know more about the person who posted anything so that you can make informed decisions about whether or not to believe or trust them based on knowing who they are, where they are, how many times they've changed their profile, or if their profile is three days old or three years old. But I don't think we should be making value judgments based on what tool you used.

28:21

SPEAKER_00

Is there an AI content creator you love that you're just like, "That's so good. I love watching these AI videos"? Yeah, what is she called? Plastic Dream Sequence? Is that what it is? [SPEAKER_01] Check it out. Plastic Dream. I have it on my phone. I'll double check. It's these dolls, Barbies, but they're singing songs and these little tiny silhouettes and snippets. It's just amazing. It's weird, but also kind of amazing. And it's very clearly AI. It's not pretending not to be. But it has a very clear creative and aesthetic point of view. And every time I come by one, I'm like, "Yep, we're doing this now. I'm going to watch this for 30 seconds."

28:34

SPEAKER_00

I have it pulled up here, and I don't—I want to watch it, but I'm not going to. [SPEAKER_01] That's awesome. If only that AI—that's another one. He's out of—he's in France, I think he's in Paris. He uses multiple different tools and models, but he tries to create these dreamscapes and animate them. So he uses one model to create the image, another one to create the video, music, etc. He's like very clearly got his own aesthetic. You could think of him as a painter, but this is his tool.

28:45

SPEAKER_00

Is there kind of a vision of AI versus human in the feed? Do you think it'll—you said you maybe want to market—like, how do you think about people? Are they going to be like "AI account, not an account"? How do you think about, or is that still kind of a work in progress?

28:51

SPEAKER_00

[SPEAKER_01] Maybe we'll end up in the same place, but there's a difference between marketing content and marketing accounts, and they're both useful and interesting. So if content was created with AI, I think you should be able to know that. That's hard, by the way, because we can detect that right now. But as these models get better, we might lose the ability to detect that. So we should also be very careful to be honest with you about how confident we are in our own assessment. But I think you should be able to just ask, "Hey, is this AI?" and we should be able to tell you who we think it probably is, or we're not sure, or it's definitely not, or definitely is. I actually think it might be more practical to label camera-captured content, basically non-AI content, as opposed to labeling AI content long-term for a couple reasons. But then at the account level, I think it also matters. There is definitely a new spam vector, which is these fake accounts. By the way, an AI creator—that's fine. There's nothing wrong with that necessarily. But there are these spam vectors which are trying to abuse that. You know, they're selling bogus supplements, and it's like an AI monk, and it doesn't present it. It's not obvious that it's AI, and it's just trying to take advantage of a certain aesthetic or stereotype. We need to figure out how to crack down on that. And so I do think we should be making sure that you just need to know, and then you can make your own informed decision. Is the account a real person or not? Is the content a real piece of content or not?

28:55

SPEAKER_00

When you think about other platforms in the space, social, you know, content platforms—are there any features or ways of approaching stuff that they do well that you're kind of jealous of or really impressed by?

28:59

SPEAKER_01

Yeah, there's a bunch. Everybody does so much. Because for me— an AI and it's just trying to take advantage of a certain aesthetic or a certain stereotype that we need to figure out how to crack down on that and so I do think we should be making sure that you just need to know and then you can make your own informed decision is the account a real person or not is the content a real piece of content or not when you [SPEAKER_00] think about other platforms in the space social content platforms are there any features or ways of approaching stuff that they do well that you're impressed by

29:20

SPEAKER_01

[SPEAKER_00] yeah there's a bunch everybody so many people do so much because for me one of the things that we are finally catching up with but I've been always very impressed with is TikTok and their recommenders ability to break small talent in the world of ranking recommenders and rank you can talk about exploitation based ranking that sounds terrible but it just means using the data you have and then you can talk about exploration based ranking going and trying to figure out what someone might be interested in that they might even not know they're interested in yet and it is much easier to move engagement by showing people stuff that you know they'll probably like because lots of people like it it's much harder to go and figure out how to essentially test content so that we can see hey maybe you sure you like Bieber but you might also like Afropunk and so we're just going to show you some Afropunk and see what happens if you do the latter this exploration based ranking you can i think it's really good for niche creators and small creators because you give them a chance to find an audience that either wasn't going to see them before or didn't even know that they were interested before so we've invested a lot over the last couple years in ranking not just increasing engagement but increasing originality increasing the number of pieces of content that break out increasing recency to stay culturally relevant and so a lot of that has been inspired by TikTok and by dance i think we're catching up there's actually a couple of those areas where I think by the best we can tell we're ahead of them there's a couple where we're still behind but we have line of sight to I think being the best in class at recommendations for the first time during my tenure so that's the I think and they get a lot of credit for inspiring a lot of that work

29:26

SPEAKER_00

nice job well we'll see not there yet they call me disappointed dad my team is always like can you ease up on the disappointed dad vibe so I'm trying to I'm trying to be a little bit more generous about giving people their flowers like you say that but that's an interesting common thread across really successful leaders is just never being satisfied yeah always feeling blessing and a curse here's all the problem it is yeah

29:52

SPEAKER_00

on this creator piece I think that's also you know people complain about this global algorithm not showing them all their friends but I feel like this is a benefit of what happens when you do this now that you can break new creators into a wide audience if you have this kind of global algorithmic feed which is really great for a lot of people I mean I'm out there talking about a lot of these contentious issues and I get beat up a lot in the comments which is fine my main thing here is

29:59

SPEAKER_00

[SPEAKER_01] just to try to communicate that there's almost always trade-offs right there's that you know you can't just have all of the things unfortunately you know you want to have you know you want to never see something you're not interested in then you're also just going to see the most basic general lowest common denominator stuff all the time you know you want to discover new and interesting things you're occasionally going to see stuff that was a mess you know but this isn't just true about ranking these all these major debates have trade-offs right you know privacy and safety those two things are in tension you know do you want a company scanning your messages or not there's some really significant trade-offs on both sides of that debate and so generally speaking when I argue and engage in debate with people who feel really strongly about things I'm not usually trying to convince them their mind is usually made up I'm just trying to enumerate all of the different pros and cons for the rest of the people watching the conversation

30:05

SPEAKER_00

speaking of getting torn apart in the comments you're so in the middle and think of all of these really hairy situations changing the feed you're in the Cambridge Analytica lawsuit all this just you're in the center of so much controversy and is that something yep oh man is that just you I will lean into this this is the thing I need to do or is it Zuck being like Adam you got to be the front face of all the stuff and get in there like where does that come from

30:09

SPEAKER_00

[SPEAKER_01] it started on newsfeed so I used to run newsfeed at Facebook and my take was that the debate was going to happen with or without us so we might as well participate and so I started being really active on Twitter specifically because that's where journalists really lived at the time and I thought it would show some humility to show up on their turf so to speak my Twitter ended up being the darkest place in my life because I just followed all of our biggest critics that's not a dig on Twitter that was just what I did and that's where it started and it slowly built from there for better for worse we've become a really important part of daily life for a lot of people we touch a lot of people we have a lot of responsibility and there's a lot of change and with change means there's going to be anxiety and stress and scrutiny we've made great decisions we've made mistakes we've been criticized for things that I think we've been criticized unfairly we've been criticized fairly and so we just need to accept that this debate is going to happen broadly so I just think it's better for us to talk about it and just be clear about what we're doing why we're doing it what the trade-offs are if people disagree that's okay we're not necessarily winning over friends when we talk about what we do but I think over the long run people are fundamentally more afraid of things they don't understand and of things where people are more secretive and less accessible and so I've tried to show up in an accessible and authentic way and I've made mistakes and I have enjoyed it at times and hated it at other times that's how it started there was also a fun debate in Mark's senior leadership team a long time ago where we were just talking about how we're a social media company where we had a very conventional approach to communication and press releases and say why don't we just use our platform so I was not in that debate but I stuck myself into that debate trying to mediate it and I think but that was also a reason why I ended up getting sucked in because Mark was like all right well let's see why don't you try and see how it goes

30:15

SPEAKER_00

what's something that helps you deal with the hate that flows at you every time you say something that people disagree with [SPEAKER_01] you try to put it in perspective right you know so it started with I did the redesign of newsfeed in 2009 we launched it March [SPEAKER_01] talking about how we're a social media company where we had a very conventional approach to communication and press releases and say why don't we just use our platform so I was not in that debate but I stuck myself into that debate trying to mediate it and I think but that was also a reason why I ended up getting sucked in because Mark was like all right well let's see

30:33

SPEAKER_00

Why don't you try and see how it goes? What's something that helps you deal with the hate that flows at you every time you say something that people disagree with? You try to put it in perspective.

30:40

SPEAKER_00

[SPEAKER_01] Right, you know, so it started with I did the redesign of newsfeed in 2009. We launched it March of 2009 for Facebook. I was a designer, I was a front-end IC designer, entry-level designer and the first comment that came in was something pretty derogatory. I think it was homophobic and anti-semitic. We're all sitting there, we launched this thing and we're just looking at the stream of comments and it's the first one and it was specifically about they don't know me but it was like, expletive expletive designed this and I was like oh it was me and I was devastated. I was a 25 year old kid and I thought about it and I came to this idea that if you spent 30, 40, 50 minutes a day at a desk and you organize your photos there and you brought letters to your friends there and you read there and then I just came and I rearranged your desk and I didn't tell you, I didn't warn you, I didn't even explain why, you would be pissed and that would be reasonable and that was what was happening with millions of people. So I try to put things in perspective and then I try to step away from it, get time with my kids, get time outside. There are months where it's really not hard at all and there are months where it's really grinds on me.

30:44

SPEAKER_00

Along those lines, there's a famous kind of reversal when you redesign the feed into this kind of video scrolly experience. There's this whole protest. The world protested, yeah, that was pretty rough. What was kind of like okay, wow, we're actually not right and we should go back? What helped you decide okay, let's change course?

30:49

SPEAKER_00

[SPEAKER_01] So actually that one, three or four things got conflated. We had a redesign of feed that went to the video viewer that was a test to four percent of users on iOS. It was not going to roll out. It was an early test to get some sense and feedback on the idea. We were also leaning into reels a lot. We were also leaning into recommendations so posts from accounts you don't follow a lot and there were also creators who were upset about the fact that their reach was going down and they were blaming ranking changes on that. Those four things got all conflated. We had some pretty big name creators publicly slap us. Then the press covered that creator sort of backlash which then got more creators doing it. So we ended up with this little bit of a multiplier effect or echo between the creator community and the press back and forth but we were never going to launch that. That was an early test. We were all, we knew it was going to need to work. We actually have continued to grow video and invest in creative tools and invest in ranking and investment recommendations and that's driven most of our growth in the years since. So I think we were pushing. I don't know what it feels like but I think we were pushing things a little bit too fast and when you are responsible for a platform like Instagram you need to be reasonable and realistic about how much you can evolve it. Now I would much rather have backlashes like that every couple years and continue to evolve and continue to stay relevant than the alternative which would have been like we didn't have video, we didn't have DMs, we didn't have stories, we didn't have ranking and we wouldn't be having this podcast right now. But the cost of leaning in is that you're occasionally going to make a mistake and you're going to definitely pay for it.

30:56

SPEAKER_01

[SPEAKER_00] It's interesting how running experiments now is really risky for companies at your scale. One person spots it and you kind of need to have a press strategy. You don't need to be proactive about communicating it but you need to have a calm strategy.

31:03

SPEAKER_01

We can't, for any design change or any test that could be controversial, we talk about it beforehand and be like okay, not if it leaks, when it leaks, what are we saying? You know, are we, should we talk about proactively? Should we talk about reactively? Either way, what's the message? Because you can't launch something to three billion people and not test it first but you can't test something at our scale and not expect people to cover it and so you have to be ready to talk about it before you even know you want to launch it. So it makes the development cycle more complicated than it used to be.

31:13

SPEAKER_01

[SPEAKER_00] Yeah, yeah. Head of growth at Anthropic launched an experiment with pricing and it just went crazy on Twitter. He's like, it's all about one percent of people, we're just trying stuff. Pricing particularly that one is a real, you got to be real careful with that one. I've learned. We've all learned these lessons. We should all share notes more. That's what I think.

31:24

SPEAKER_01

[SPEAKER_00] There's one not to do. How to avoid the internet hating you for the day. Yeah, yeah, I'm happy to talk to that. I'm both in. I think he's all right. He's all right. Okay, I'm going to take us to two recurring corners on the podcast: fail corner and hot seat corner. Fail corner, what's something that you worked on that was just a huge failure that helped you become better?

31:32

SPEAKER_01

Oh, a bunch. So I'll give you two, maybe. So before Instagram, my first project as a PM was on a project called Facebook Home, which was a sort of fork of Android at the operating system level and a piece of hardware with HTC. It was a spectacular failure. I learned way more in that year and a half than I did any year, probably in my career, because I was a design manager before that. I declared myself a PM because the PM on the project quit and I just threw myself head first into understanding carriers and OEMs and certification as well as Android and operating systems and just learned a ton. So I'm happy I brought that project to an end because it had been going on for a long time and sometimes the best thing you can do is execute an idea that doesn't have market fit just to decide whether or not the idea was a good idea in the first place. Another big mistake I made under my Instagram tenure was the first version of reels was built on top of stories. Stories had a ton of momentum. This was I think 2019 and we were trying to build reels into stories because we were trying to build on the thing that was growing the fastest but it was not a strong foundation. The read-through rate on stories is relatively low. There are way more stories than most people have.

31:38

SPEAKER_01

I lend a ton, and I'm happy I brought that project to an end because it had been going on for a long time. Sometimes the best thing you can do is execute an idea that doesn't have market fit, just to decide whether or not the idea was a good idea in the first place. Another big mistake I made under my Instagram tenure was the first version of reels was built on top of stories. Stories had a ton of momentum. This was, I think, 2019, and we were trying to build reels into stories because we were trying to build on the thing that was growing the fastest, but it was not a strong foundation. The read-through rate on stories is relatively low. There are way more stories than most people have time to consume, so most of the reels were never seen and then they disappeared. If we had the version of reels that we launched in mid-2020, maybe the summer of 2020 in the summer of 2019, I don't think TikTok isn't—I think TikTok is still big and important, but I don't think it's as big as it is now. When they really took off was when the pandemic hit and a bunch of people had a lot of time at home and were looking for joy and were totally fine with their phone having sound on. If you look at the numbers, 2020 is when they exploded, and we were out of position. On one hand, I'm a designer. I'm trying to not add new things to the product. I'm trying to extend existing primitives, and that was the idea. On the other hand, that was wrong, and it's a pretty big fork in the road if you just look at the overall business over the last eight years. We create a lot of economic opportunity in the world, allowing TikTok to grow. So there's a lot I'm glad they exist.

31:45

SPEAKER_01

[SPEAKER_00] Okay, final question. I'm curious about your screen time policy with your kids. I know you have three kids. There's a lot of concern these days about Instagram not being great for kids. A lot of tech executives don't let their kids use devices while they're building the product. As head of Instagram, how do you think about screen time with your kids?

31:54

SPEAKER_01

The key thing for me is boundaries. It's also about education and having conversations with them. My kids are too young to use social media. They're 10, 8, and 6, but they each have an iPad. They have to earn their time. They have different ways they earn that time. It's usually about sitting down to do your homework three times for half an hour each, which gets you a total of 90 minutes on the weekend. Then they can use that time on the weekend. But you have to set that boundary where it's like you can't just ask for it and you give it to them. I think that matters a lot. I'm pretty opinionated about what they do on it. I approve what apps they have. I think parents should be approving what apps kids are downloading onto their devices. We've been advocating for this at a policy level for a long time at Meta. I think those things help a lot. There are some exceptions. One is planes. It's just about surviving. I don't know if you've ever—for those of you who are parents, I'm going to say yeah, it's like you just need to get through it. It's a 10-hour flight or 8-hour flight. You just need to get through it. The other one that I'm starting to experiment with my 10-year-old with is school. Schools are interesting because I think I'm pretty supportive of no phones in classrooms. That's happening more and more. I think that's probably good for education. I also know in the world of AI that there's concern about kids using AI and not learning critical thinking skills. I think that's a valid concern, but I'm also worried about kids not learning how to leverage AI and then being at a disadvantage. That's a balance. I think you need both. With my eldest, we started doing Replit coding recently together. He loves video games, so I was like all right, let's make a video game. He's made this 19-level platformer game that looks like an 8-bit version of Super Mario from when I was a kid, but each level has its own theme, its own types of monsters. There's a store where you can buy different skins or weapons. It's unbelievable what a 10-year-old who still types with three fingers can do with just a couple hours of sitting and doing it together. But that is more of like I want you to learn how to make things. I want you to be thinking, not just playing games. I'm going to sit with you and do that. We're going to do this together. To me, these are the things that matter: boundaries, scoping it down to the activities you think are healthy for your kid. Every kid is different, but I do think you want your kids to be digitally literate and AI literate because I think if they're not, they're going to be at a disadvantage. But you also don't want it to be a free-for-all.

31:59

SPEAKER_01

[SPEAKER_00] This is selfishly useful for me as someone with a three-year-old, and I'm trying to figure all this stuff out. So this is useful. For me to figure out a strategy, it's amazing. It's a thing, and you're not that far off. You're really not that far off. It's going to happen in a couple years. Replit coding next year. Let's do it. I couldn't believe I tried to do it six months ago and it just totally didn't work, and then now with the new models it's been amazing.

32:15

SPEAKER_00

What's their platform of choice? Are they using Replit coding? [SPEAKER_01] Yeah, my 10-year-old is using Replit right now. Amazing. But we will see how that goes. Adam, I'm going to let you go. Thank you so much for being here. You're just such a gem of a person. It's so obvious how clear and authentic you are and how deeply you think about everything. I really appreciate you being here.

32:33

SPEAKER_01

I appreciate you bringing me on. I've been a fan for a long time. It's nice to finally get to have a conversation. I really appreciate that. [SPEAKER_00] Let me just ask you this final question. I ask everyone: what's the way that listeners can be useful to you?

32:47

SPEAKER_01

I just think you don't even have to tell this to other people, but just remember that this world and technology is complicated, and there are almost always trade-offs. You can totally disagree with the decisions I or we make, but just remember that we are people here trying to make these decisions, trying to do the best we can. I actually do invite the criticism, the critique, and the feedback. But know that none of these contentious debates are nearly as simple as most people pretend to make them out to be. [SPEAKER_00] Wise words, Adam. Thank you so much for being here. Pleasure.

33:03

SPEAKER_01

[SPEAKER_00] Thank you, Lenny. Bye, everyone. Thank you so much for listening. If you found this valuable, you can subscribe to the show on Apple Podcasts, Spotify, or your favorite podcast app. Also, please consider giving us a rating or leaving a review, as that really helps other listeners find the podcast. You can find all past episodes or learn more about the show at lennys.podcast.com. See you in the next episode. those five complement each other i need to make and that's you know about what skills each one has what weaknesses each one might have i also need to make sure that they um this is more art than science

33:22

SPEAKER_01

have a good vibe right you know you need you know trust and rapport a leadership team with strong trust and rapport can work through most anything a leadership team without trust or rapport like anything can become an issue and so that that chemistry bit is is is like i said much more art than science but that also matters and so i think some of the best leaders and product leaders specifically also either do that instinctively or consciously but you know they haven't they have a nose for for building teams that are going to have good energy and um good collaboration warm and fuzzy stuff

34:00

SPEAKER_01

yeah yeah well this the flip that is also true right like i've had many times in my career i've had two people who i think are amazing and i even adore them and love them and they just can't get along you're like this isn't a competency issue this is just a personality issue and you just have to

34:20

SPEAKER_00

to sometimes call it and split them i want to transition talk to talk about instagram the product the platform things you guys have learned there let me start with this question what's something that the instagram algorithm knows about human behavior that people may not realize one of the most common

34:39

SPEAKER_01

misconceptions is actually in the opposite direction i think people assume that there's a much more detailed semantic understanding of everybody's interests and preferences in the algorithm than there is most of what's really driven the progress in the world of recommenders over the last five ten years have been you know these large embedding models and these other techniques that basically produce artifacts that cannot be read by people they're not legible they're like giant vectors it's like sure i can show you the vector but it's just going to be a bunch of numbers in like a

35:12

SPEAKER_01

seven-dimensional space it's like and so when when we talk about does the algorithm know something usually we think in these more semantic terms it knows i like surfing and it's like it doesn't it just has this big-ass number that happens to correlate with surfing um that said i think that is starting to change right i think that what one of the things that lms are enabling is they can describe in you know words you know english for or whatever language you prefer what some of those previously illegible artifacts um are at least proximate to if not mean directly right so this is like the thing i've been really i posted about this

35:59

SPEAKER_01

this week this thing called your algorithm basically the idea is we take a look at all of the stuff that you've interacted with and then you know we all of that is in an embedding space you can think of embedding space as a map you can map a bunch of videos into the same map and so videos that are close are similar and now we can just have an lm just be like describe that part of the map and it can be like oh that is like deep pour over coffee snobbery and um and that's kind of amazing that is so cool

36:30

SPEAKER_00

that like you can ask the lm to look at these numbers and extrapolate here's like the topic that

36:35

SPEAKER_01

you're interested in yeah or look at the videos and so the way you um both and so you can also embed concepts into that same space and so i mean embeddings are really the underlying technology underneath llms right that's how they all the whole thing works and so you know so what what we what we what we do now is we let you you know quote unquote see your algorithm you can see what topics we think you're interested in um and you can adjust it you can add and remove things but the idea here giving people some agency back in a world where you know these social media apps are getting taken over

37:11

SPEAKER_01

by recommendations but we can't do a lot of other things yet which we will be able to do you know you could there's things that aren't topical that you might ask for i want more fun content i want to see my friends more i don't want to see my high school's kids friends kids photos you know i don't know we can come up with it i don't want to see seven photos in a row but i'm happy to see six photos or whatever your hearts can you know whatever your mind can come up with so we have a lot of work to do and so i'm excited about that but um i think a misconception historically is until recently

37:43

SPEAKER_01

we don't really know as much about you as you think we're just like oh like you liked these photos this these people also like those same photos and they like these other photos so you might like those other photos like that's kind of how i'm oversimplifying that's like yeah kind of how it worked now only now are we actually getting as sophisticated as i think people have assumed we've been for many

38:05

SPEAKER_00

years that is really interesting one that comes to mind is kind of this transition everyone eventually goes through to this like algorithmic broad global feed uh everyone it always feels like people think i just want to see chronologically everyone i know and follow and that's going to be my favorite feed and it continues to be proven wrong no you actually engage a lot more a lot more when it's this

38:28

SPEAKER_01

algorithmic feed of things we think you will love yeah it's tough because i mean i get i mean i posted this week this thing about agency and i just got destroyed in the comments which is just part of the job i get it right um there's but there are a couple issues with the algorithm with the chronological feed so one is and some of this is the tension between an individual's interests and what works when you scale it up right so if you do a pure chronological feed the incentive for everybody is to just post as much as possible because it will always be at the top of everyone who follows you's

39:08

SPEAKER_01

feed as soon as you post so what ends up happening is that the fee gets overwhelmed with professional content with usually large company content and publishers because they get you know the new york times can pump out 50 things a day your your best friend won't you know you might get one thing a week from them and so your feed just gets taken over so part of it is the incentives that emerge because when when you design these systems it's almost like designing a city you need to think about okay here's the here the here's how the mechanics work what are the incentives that arise how are people

39:42

SPEAKER_01

going to act within those incentives and then what happens and the other thing is sometimes the most interesting thing was just not the most recent thing recency is an important input into relevance but it's not the only one my sister got engaged last night and you know she's in germany but she didn't if

39:58

SPEAKER_00

she did she's married she got married last year that's why it was tough in mind but if she got

40:03

SPEAKER_01

engaged you know and i missed it because she lives in europe and you know with different time differences like do i really want to see a picture of like my brother's pobo sandwich you know po boy sandwich or do i want to like see my sister's things first so i i it's tough it's tough i'd love to figure out a way to find the right balance i want to give people agency over the experience but i think it needs to be in a way that creates a system that makes sense not just for us as a business which matters i'm not pretending that's not an issue but also for the overall community because we've done chronological

40:35

SPEAKER_01

by default and where you can make it default and you see not only does usage go down overall sentiment goes down the individual who made that choice might be happy at the moment but when you just get pummeled with stuff you're less interested in over the course of months we ask we run surveys a massive scales we just see people start to become less and less satisfied with instagram

40:56

SPEAKER_00

kind of along these lines uh everybody asks you about this these days uh ai and content and how that all impacts everything that's going on i want to ask you something i haven't seen someone ask you is the rise of ai content uh a headwind or a tailwind for instagram versus other platforms do you

41:14

SPEAKER_01

think this helps or hurts you guys i think it's going to be a tailwind but i think it's going to be a challenge it's and not just because it's more content obviously we're an attention business driven business we're an advertising business more content means potentially more attention that's not for free though like i don't think we're very good at ranking ai content yet this great ai content this crap ai content you should just see the stuff you're interested in and not any of the stuff you're not interested in but i do think that in a world where or for years now and i've said this many times power is shifting from institutions to individuals across industries

41:48

SPEAKER_01

the easiest example of sports where players are more relevant than teams now and that was not the case when i was a kid in that world you're gonna i think it behooves us to invest in individuals and to invest in specifically for instagram and creators and i mean creators broadly i don't just mean influencers who are promoting branded content and making you know native only videos i mean anybody who's using platforms like instagram to help do what they do right it could be you could be a journalist you could be an artist you could be selling scarves you sew but like you're out there as yourself

42:27

SPEAKER_01

creating and sharing content that helps you achieve whatever it is you're trying to do so we've been leaning in that direction for many years now that's been our you know one of our two or three most important audiences for as long as i've been on instagram in a world where there's an abundance of synthetic content i actually think people are going to seek out creativity and authenticity and people more not less and i think that that will help us that doesn't mean that we won't have ai content on our platform there's going to be bad and good ai content and we're going to try and handle that you

43:00

SPEAKER_01

know the way we normally handle content so unsafe goes away interesting versus not interesting is based on ranking and personalization but i think people are going to really seek out other points of view because instagram was never just about the content it was always about to a certain degree the person behind the content the point of view the reason they're sharing it their perspective and i think that's going to become more important not less and i think given that we are not the best at a lot of things but we are the largest uh creator platform you know you know if you look at how we define creators

43:35

SPEAKER_01

and how many creators use us versus other platforms i think it'll be a tailwind for us because i think

43:40

SPEAKER_00

people are going to seek out people and this connects to your earlier point that uh companies uh like say new york times can pump out a bunch of ai content versus a creator and so you're saying you kind of want to protect against that to allow individuals to continue to perform well in spite of just all

43:57

SPEAKER_01

aic content if you just love ai content great like you should be able to have a feed that's just like ai town and if you don't then you shouldn't have it in your feed like you know to me it's like i don't think we should i mean i understand why people are right there's a i'm not oblivious to the overall paradigm shift and sort of revolution that we're sitting in but i don't think we should judge content based on the tool that made it um i think we should judge it based on the content the point of view the person behind the content like i don't i don't think we should filter out ai content i think we

44:34

SPEAKER_01

should let you know if content is ai content or not i think we should let you know more about the person who posted anything so that you can make informed decisions about whether or not to believe or trust them based on you know knowing who they are or where they are or how many times they've changed their profile or you know you know if their profile is three days old or three years old um but i don't

44:56

SPEAKER_00

think we should be making value judgments based on what tool you used is there an ai uh content creator you love that you're just like that's so good about watching these ai videos yeah what is she

45:07

SPEAKER_01

called plastic plastic dream sequence is that what it is um i think check it out check it out yeah plastic dream uh um plastic dream sequence i have it on my phone i'll double check um it's these like uh like sort of dolls barbies but they're like singing songs and these little tiny silhouettes and snippets and it's just it's just amazing it's like a little weird but like also kind of amazing and it's it's very clearly ai it's not pretending not to be but it's has a very clear creative and aesthetic point of view and every time i come by one i'm like yep we're doing this now i'm gonna watch this for 30 seconds

45:47

SPEAKER_00

i have it pulled up here and i don't i want to watch it but i'm not going to

45:52

SPEAKER_01

that's awesome if only that ai that's another one he's out of he's in france i think he's in paris he uses multiple different tools and models but he kind of tries to create these dreamscapes and animate them so he create uses one model to create the image another one to create the video music etc uh he's like very clearly got his own aesthetic um uh and he's just like you could you could think of him as a painter but like this is his tool is there kind of a vision of ai versus human

46:21

SPEAKER_00

in the feed do you think it'll like you said you maybe want to market like how do you think about people are they going to be like ai account not an account how do you think about or is that still

46:29

SPEAKER_01

kind of a work in progress maybe we'll end up in the same place but there's a difference between marketing content and marketing accounts and they're both useful and interesting so if content was created with ai i think you should be able to know that uh that's hard by the way because we can detect that right now but as these models get better we might lose the ability to detect that so we should also be very careful to be honest with you about how confident we are in our own sort of assessment but i think you should be able to just ask be like hey is this ai and we should be

47:00

SPEAKER_01

able to tell you who we think it probably is or we're not sure or it's definitely not or definitely is um i actually think we might be might be more practical to label camera captured content like basically non-ai content as opposed to labeling ai content long term for a couple reasons but then at the account level i think it also matters there is definitely an a new spam vector which is these fake accounts which by the way an ai creator that's fine there's nothing wrong with that necessarily but there is there are these spam vectors which are trying to abuse that and you know they're selling like

47:35

SPEAKER_01

you know bogus supplements and it's like an ai monk and it doesn't present it it's not obvious that it's an ai and it's just trying to like take advantage of you know a certain aesthetic or a certain sort of stereotype that we need to figure out how to crack down on that and so i do think i do think we should be making sure that you know basically you just need to know and then you can make your own informed decision is the account a real person or not is the content a real piece of content or not when you

48:01

SPEAKER_00

think about other platforms uh in the space social you know content platforms are there any um features or just or like ways of of approaching stuff that they do well that you're kind of jealous of or really impressed by yeah there's a bunch everybody so many people do so much because i mean for me

48:19

SPEAKER_01

like one of the things that we are finally catching up with but i've been always very impressed with is tick tock and their recommenders ability to break small talent in the world of ranking recommenders and rank you can talk about exploitation based ranking that sounds terrible but it just means like using the data you have and then you can talk about exploration based ranking going and trying to figure out you know what someone might be interested in that they might even not know they're interested in yet and it is much easier to move engagement by showing people stuff that you know they'll probably

48:54

SPEAKER_01

like because lots of people like it it's much harder to go and figure out how to essentially test content so that we can see like hey maybe you sure you like bieber but you might also like afropunk and so we're just going to like show you some afropunk and see what happens if you do the latter this exploration based ranking you can i think it's really good for niche creators and small creators because you give them a chance to find an audience that either wasn't going to see them before or didn't even know that they were interested before so we've invested a lot over the last couple years

49:29

SPEAKER_01

in ranking not just increasing engagement but increasing originality increasing the number of pieces of content that break out increasing recency to stay culturally relevant and so a lot of that has been inspired by by tick tock and by dance i think we're catching up there's actually a couple of those areas where we i think by the best we can tell we're ahead of them there's a couple where we're still behind but for we have line of sight to i think being the best in class at recommendations for the first time um during my tenure uh so that's the i think and they get a lot of credit for inspiring

50:03

SPEAKER_01

a lot of that work nice job well we'll see not there yet they call me disappointed dad my team is always like can you ease up on the disappointed dad vibe so i'm trying to i'm trying to be a little bit more um generous about giving people their flowers like you say that but that's an interesting common

50:24

SPEAKER_00

thread across really successful leaders is just never being satisfied yeah always feeling like blessing and a curse here's all the problem it is yeah on this creator piece i think that's also you know people uh complain about this global algorithm not showing them all their friends but i feel like this is a benefit of what happens when you do this now that you can break new creators into a wide audience if you have this kind of global algorithmic feed which is really great for a lot of people i mean i'm out there talking about a lot of these contentious issues and i get beat up a lot in the comments which is fine my main thing here is

51:02

SPEAKER_01

just to try to communicate that there's almost always trade-offs right there's that you know you can't just have all of the things unfortunately you know you want to have um you know you want to never see something you're not interested in then you're also just going to see like the most basic general lowest common denominator stuff all the time you know you want to discover new and interesting things you're occasionally going to see stuff that was just a mess you know you know but this isn't just true about ranking these all all these major debates have trade-offs right you know uh privacy and

51:38

SPEAKER_01

safety those two things are intention you know do you do you want a company scanning your messages or not there's some really significant trade-offs on both sides of that debate um and so generally speaking when i argue and engage in debate with people who feel really strongly about things i'm not usually trying to convince them they usually their mind is usually made up i'm just trying to enumerate all of the different puts and takes for the rest of the people watching the conversation

52:05

SPEAKER_00

speaking of getting torn apart in the comments like you're so in the middle and think of all of these really hairy situations changing the feed you're like in the cambridge analytica lawsuit all this just like you're in the center of so much controversy and uh is that something yep oh man is that just like you i will lean into this this is the thing i need to do or is it like zuck being like adam you got to be the front face of all the stuff and get in there like where does that come

52:37

SPEAKER_01

from it started on newsfeed so i used to run newsfeed at facebook and i my take was that the debate was going to happen with or without us so we might as well participate and so i started being really active on twitter specifically because that's where journalists really lived at the time and i thought it'd be show some humility to show up on their turf so to speak my twitter ended up being like the most the darkest place in my life because i just followed all of our biggest critics that's not a dig on twitter that was just like what i did um and that's where it started and it kind of slowly

53:13

SPEAKER_01

built from there for better for worse we've become a really important part of daily life for a lot of people we touch a lot of people we have a lot of responsibility and there's a lot of change and there's with change means there's going to be anxiety and stress and scrutiny we've made great decisions we've made mistakes um we've been criticized for things that i think um we've been criticized unfairly we've been criticized fairly and so we just need to accept that this debate is going to happen broadly so i just think it's better for us to talk about it um and just be clear about what we're doing why we're doing it

53:53

SPEAKER_01

what the trade-offs are if people disagree that's okay we're not necessarily you know winning over friends when we talk about what we do but i think over the long run people are fundamentally more afraid of things they don't understand um and about things where people are more secretive and less accessible and so i've been tried i've tried to show up in an accessible and authentic way um and i've made mistakes and i have enjoyed it at times and hated it at other times um that's kind of how it started there was also kind of a fun debate in mark's sort of senior leadership team a long time ago where we were just

54:28

SPEAKER_01

talking about how we're a social media company where we had like a very sort of conventional approach to communication and like press releases and say why don't we just use our platform so um i was not in that debate but i stuck myself into that debate trying to mediate it and i think um but that was also a reason why i ended up getting sucked in because mark was like all right well let's see

54:46

SPEAKER_00

like why don't you try and see how it goes what's something that helps you deal with the the hate that flows at you every time you say something that people disagree with you try to put it in perspective

54:57

SPEAKER_01

right you know like so it started with i did the redesign of newsfeed in 2009 we launched it march of 2009 for facebook i was a designer i was a front like an ic designer front you know entry-level designer and the first comment that came in was something pretty derogatory i think it was like it was like homophobic and anti-semitic you know it was just like literally we're all sitting there we launched this thing and we're just looking at the stream of comments and it's like the first one and it was specifically about they don't know me but it was like what expletive expletive um sensor

55:35

SPEAKER_01

sensor uh designed this and i was like oh it was me um and i was like devastated i was like 25 year old kid and i don't know i thought about it and i came i came to this idea that if you spent 30 40 50 minutes a day at a at your desk and you organize your photos there and you brought letters to your friends there and you read there and then i just came and i rearranged your desk and i didn't tell you i didn't warn you i didn't even explain why like you would be pissed and that would be reasonable um and that was what was happening just you know with millions of people um so i try to put things in perspective

56:19

SPEAKER_01

and then i try to step away from it um get time with my kids get time outside there there are months where it's really not hard at all and there are months where it's

56:28

SPEAKER_00

really really grinds on me along those lines there's a famous kind of reversal when you redesign the feed into this kind of video scrolly experience there's this whole protest the world protested yeah that was pretty rough uh what was kind of like okay wow we're actually not right and we should go back what was kind of what helped you decide okay let's change course so actually that one

56:51

SPEAKER_01

got that one three or four things got conflated we had a redesign of feed that went to the video viewer that was a test to four percent of users on ios it was a not it was not going to roll out it was this like an early test to get some sense and feedback on the idea we were also leaning into reels a lot we were also leaning into recommendations so posts from accounts you don't follow a lot and there were also creators who were upset about the fact that their reach was going down and they were blaming ranking changes on that those four things got all conflated we had some pretty big name creators publicly like

57:28

SPEAKER_01

slap us then the press covered that crater sort of backlash which then got more creators doing it so we ended up with this little bit of like a multiplier effect or echo between the creator community and the press back and forth but we were never going to launch that that was an early test we were all we knew it was going to need to work we actually have continued to grow video and invest in creative tools and invest in ranking and investment recommendations and that's driven in the most of our growth in the years since um so i think we were pushing well i don't know what does it feel like but like it i think

58:08

SPEAKER_01

we were i think my real takeaway wasn't that we should have not tested that design necessarily i think we could have been we could have done a bunch of things better to explain and maybe move a little fast move a little slower i think we were just pushing things a little bit too fast and when you are responsible for a platform like instagram you need to be reasonable and realistic about how much you can evolve it now i would much rather have backlashes like that every couple years or continue to evolve and continue to stay relevant than the alternative which would have been like we didn't have video we

58:41

SPEAKER_01

didn't have dms we didn't have stories we didn't have ranking and we wouldn't be on having this podcast right now um um but the cost of leaning in is that you're gonna occasionally like make a mistake and

58:54

SPEAKER_00

you're gonna definitely um pay for it it's interesting how running experiments now is like very risky for companies at your scale one person spots it and i go you kind of need to have a press you don't need to

59:07

SPEAKER_01

be proactive about communicating it but you need to have a calm strategy like we can't for any for any design change or any test that could be controversial we we talk about it beforehand and be like okay not if it leaks when it leaks what are we saying you know are we you know should we talk about proactively should we talk about reactivity either way what's the message because you can't you can't you can't you can't launch something to three billion people and not test it first but you can't test something at our scale and not expect people to cover it and not and be and so you have to be ready to talk about

59:44

SPEAKER_01

it before you even know you want to launch it um so it's um it makes the development cycle more

59:52

SPEAKER_00

complicated than it used to be yeah uh yeah head of growth at anthropic launched an experiment with pricing and it just went crazy on twitter he's like it's all about one percent of people were just

1:00:01

SPEAKER_01

trying stuff like no pricing particularly that one is a real you got to be real careful with that one i've i've learned we've all learned these lessons we should all share notes more that's what i think

1:00:12

SPEAKER_00

there's one not to do how to avoid the internet hating you for the day yeah yeah i'm happy to talk to that i'm both in a thrombing i think he's all right he's all right okay i'm going to take us to two recurring corners on the podcast fail corner and hot seat corner fail corner what's uh what's something that you worked on that was just a huge failure that helped you become better oh a bunch um so

1:00:35

SPEAKER_01

any so i'll give you two maybe so before instagram my first project as a pm was on a project called facebook home which was a sort of fork of android um at the operating system level and a piece of hardware with htc it was a spectacular failure i learned way more in that year year and a half and i did it any year i think probably in my career because i was just a design manager before that i declared myself a pm because the pm on the project quit and i just threw myself head first and understanding carriers and oems and uh certification as well as android and operating systems and just

1:01:15

SPEAKER_01

lend a ton um so and i'm happy i brought that project to an end because it had been going on for a long time and sometimes you the best thing you can do is execute an idea that doesn't have market fit well just to decide whether or not the idea was a good idea in the first place another big mistake i made under my instagram tenure was the first version of reels was built on top of stories stories had a ton of momentum this was i think 2019 and we were trying to build reels into stories because we were trying to build on the thing that was growing the fastest but it was not a strong foundation most you know the

1:01:55

SPEAKER_01

read-through rate on stories is relatively low there's way more stories than most people have time to consume so most of the reels were never seen and then they disappeared um and if we had the version of reels that we launched in like mid just maybe we think it's like the summer of 2020 in the summer of 2019 i think i don't think tick tock isn't i think tick tock is still big and important but i don't think it's as big as it is now because when they really took off was when the pandemic hit and a bunch of people had a lot of time at home and were looking for a little bit of joy and we're totally

1:02:29

SPEAKER_01

fine with our phone having sound on and so if you look at the numbers the 2020 is when they exploded and we were out of position and um on one hand you know i'm a designer i'm trying to not add new things to the product i'm trying to extend existing primitives and that was the idea on the other hand that was wrong and and it's a pretty big fork in the road if you just look at the overall business

1:02:53

SPEAKER_00

over the last eight years we create a lot of economic opportunity in the world allowing a tick tock to to grow so there's a lot i'm glad they exist okay final question um i'm curious just about your screen time policy with your kids i know you have three kids uh there's a lot of concern these days about instagram for not being great for people not for kids a lot of tech executives don't let their kids use devices while they're building the product as head of instagram how do you think

1:03:25

SPEAKER_01

about screen time with your kids the key thing for me is boundaries um it's also about education and being and having conversations with them but my kids are too young to use social media they're 10 8 and 6 but they each have an ipad they get um they have to earn their time um so they have different ways they earn that time it's usually about like sitting down to do your homework three times for half an hour each gets you the total of 90 minutes on the weekend and then they can use that time on the weekend but you kind of have to set that boundary where it's like you can't just like you know we can't just be

1:04:01

SPEAKER_01

they ask for it and you give it to them i think that matters a lot and then i'm pretty opinionated about what they do on it like i approve what apps that they have i think parents should be approving what apps to kids specifically are um downloading onto their devices we've been advocating for this at a policy level for a long time i metta i think those things help a lot um there are some exceptions um one is planes it's just like about surviving i don't know if you've ever for those of you are parents i'm going to be yeah yeah it's like you just like that's just like all right you know we're

1:04:34

SPEAKER_01

you know it's a 10-hour flight or eight-hour flight it's like yeah just just you just need to get through it um the other one that i'm starting to experiment with my 10 year old with is so schools are interesting because i think i'm pretty supportive of a no phone in classrooms um that's happening more and more i think that's just probably good for education and i do also know in the world of ai that there's concern about kids using ai and not learning critical thinking skills i think that's a valid concern but i also am worried about kids not learning how to leverage ai and then being

1:05:09

SPEAKER_01

sort of at a disadvantage so that's a balance i think you need both so with with my eldest we started um vibe coding recently together um he's just loves video games so i was like all right let's make a video game and so he's made this 19 level platformer game that kind of looks like an 8-bit version of super mario from when i was a kid but like each level has its own theme its own types of monsters there's a store where you can buy different skins or weapons and there's like uh like well it's unbelievable what a 10 year old who still types with three fingers can do um with just you know a couple hours of sitting

1:05:51

SPEAKER_01

doing together but that is more of like a i want you to learn how to um make things i want you to be thinking not just playing games and i'm going to sit with you and do that we're going to do this together so to me these are the things that matter boundaries um scoping it down to the activities you think are healthy for your kid every kid is different um but i do think you want your kids to be digitally literate um ai literate because i think if they're not they're going to be at a disadvantage but just you

1:06:20

SPEAKER_00

also don't want it to be a free-for-all this is selfishly useful for me as a as a i have a three year old and i'm trying to figure all this stuff out so this is useful oh yeah no for me to figure

1:06:28

SPEAKER_01

out a strategy it's amazing it's a thing and you're you're you're not that far off you're really just

1:06:33

SPEAKER_00

not that far off it's going to happen in a couple years five coding next year let's do it i couldn't believe i tried to do it six months ago and it just totally didn't work and then now with the new remodels it's been amazing what's their platform of choice are they clock coding uh person yeah yeah my

1:06:49

SPEAKER_01

10 year old is using is using cloud code right now amazing um but um we will see we'll see how

1:06:55

SPEAKER_00

that goes adam uh i'm gonna let you go thank you so much for being here this you're just like such a gem of a person it's just so obvious how clear like how authentic you are and just like how deep leaf you think about everything uh so i really appreciate you being here i appreciate you bringing

1:07:11

SPEAKER_01

me on um i've been a fan for a long time it's nice to finally get to have a conversation i really appreciate that

1:07:17

SPEAKER_00

uh let me just ask you this final question ask everyone what's the way that listeners can be useful to you

1:07:22

SPEAKER_01

i just think you don't even have to tell this to other people but just remember that this world and technology is complicated and there are almost always trade-offs um and you can totally disagree with the decisions i or we make um but just remember that we are people here trying to make these decisions just trying to do the best we can and i actually do invite the criticism and the critique and the feedback but um but know that none of these contentious debates are nearly as simple

1:07:55

SPEAKER_00

as most people pretend to make them out to be wise words adam thank you so much for being here pleasure thank you lenny bye everyone thank you so much for listening if you found this valuable you can subscribe to the show on apple podcasts spotify or your favorite podcast app also please consider giving us a rating or leaving a review as that really helps other listeners find the podcast you can find all past episodes or learn more about the show at lenny's podcast.com see you in the next episode

Reading tools

Type to find a passage

Appearance
Ask this transcript

Notebook

Export Markdown

Add a note