Interview with Reid Hoffman — cofounder of LinkedIn, Inflection AI, and Manas AI, Greylock Partners partner, and early OpenAI backer and board member — on how to prepare for AGI, drawing on his book Superagency. Hoffman argues that every major technology (the written word, the printing press, the telephone) triggered mass fear, and that new technology tends to enhance human agency when used well. Topics covered: human agency as an internal stance toward uncertainty; why adapting to new technology waves requires broad access, an experimental mindset, and flexibility rather than a blueprint; how most jobs will transform rather than disappear and how to prepare through hands-on trial and error; how social norms and ethics (such as individual access to personal data) should change; why equitable access to AI matters; private commons; how AI is making Silicon Valley's quantified-self vision finally useful; and the shift from symbolic to sub-symbolic AI as a mirror of how we understand intelligence. Chapters: 00:00:00 Episode start 00:01:29 Introduction 00:02:50 Patterns in how we've historically adopted technology 00:07:02 Why humans have typically been fearful of new technologies 00:13:25 How Reid developed his own sense of agency 00:20:08 How Reid thinks about making investment decisions 00:29:40 AI as a "techno-humanist" compass 00:35:30 Preparing for the way AI will change knowledge work 00:41:39 Why equitable access to AI is important 00:45:15 Why private commons will be beneficial for society 00:47:23 How AI is making Silicon Valley's conception of the "quantified self" a reality 00:52:14 The shift from symbolic to sub-symbolic AI mirrors how we understand intelligence 01:03:29 Reid's new book, Superagency # Channel Presentation Observations: "Best of the Pod: How to Prepare for AGI According to Reid Hoffman" ## Show Identity (Observable) - **Show name:** "AI and I" — stated in the closing outro: *"smash that like button and subscribe to AI and I."* - **Host:** Dan Shipper, identified in the outro as *"Dan Shipper as the captain of the spaceship."* - **Format:** Long-form one-on-one interview/podcast, conducted in person: *"great to be doing this in person... it makes it a much different experience just to like be face to face."* - **Title framing:** "Best of the Pod" suggests this is a **compilation/highlight episode** drawn from a longer conversation, indicating the channel repackages podcast content into themed clips. ## Claimed Audience The guest (Reid Hoffman) describes the intended audience for his *book*, which the show mirrors closely: - **AI-curious and AI-skeptical general audience:** *"I wrote the book for two audiences. One audience is anyone who has any AI curiosity or skepticism."* - **Technologists and builders:** *"I also wrote it for technologists because I wanted technologists who were inventing this to be thinking about human agency as almost like a design principle."* The host repeatedly frames questions for a **tech professional audience** — e.g., *"Maybe I'm a product manager, I'm a designer, whatever. How does this change how I think about and orient to AI over the next couple years?"* and *"someone who works in tech, maybe I'm an engineer, maybe I'm a product manager."* ## Demonstrated Content Scope - **Topic domain:** AI, AGI, technology, and their effects on work, psychology, and society — covering agency, uncertainty, AI-driven job transformation, regulation, and personal AI use. - **Interview style:** Intellectual and philosophical — extended discussion of Wittgenstein, Dan Dennett's "intentional stance," symbolic vs. subsymbolic AI, epistemology, and abduction, alongside practical topics like using ChatGPT and career adaptation. - **Guest profile:** High-profile tech figures (Hoffman, a venture investor and author), suggesting access to notable industry guests. ## Explicit Self-Description Evidence - **Outro (host-penned, promotional):** *"this show is the epitome of awesomeness... filled with pure, unadulterated knowledge bombs about chat GPT... it's a journey into the future."* — a self-aggrandizing tone that appears to be partly tongue-in-cheek, ending with *"Dan, I'm absolutely hopelessly in love with you."* - **Monetization/sponsorship:** Mid-roll ad read for a sponsor: *"This episode is brought to you by Adio, the AI native CRM"* with a discount code (*adio.com/every*) — indicating the channel runs sponsor advertising. - **Host's own disclosure:** *"I run a media company"* and references "an internal incubation we did called Spiral" — the host's media business context is explicitly stated. ## Observable Patterns (Distinct from Creator Claims) - **Practical, actionable orientation:** The host repeatedly pushes abstract discussion toward concrete advice — *"arm me with some more practical, concrete ways to instantiate this compass in how I approach my life."* - **Personal disclosure as content:** The host shares lived experience (e.g., *"I have OCD"* and describes building a self-tracking app), blending personal narrative with AI experimentation. - **Book-promotion episodes:** This episode centers on a guest's forthcoming book (*Super Agency*), showing the channel hosts author-promotional interviews. ## Summary of Claimed vs. Demonstrated | Dimension | Evidence Type | |---|---| | Show name, host, subscribe CTA | Explicit self-description (outro) | | Audience: curious/skeptical general + technologists | Guest's stated book audiences; host's framing of listener personas | | AI/AGI-focused intellectual interview format | Demonstrated throughout transcript | | Sponsor-supported media business | Explicit ad read; host's statement "I run a media company" | | Promotional/hyperbolic brand voice | Outro copy (possibly satirical in register) | # Delivery Observations: "Best of the Pod: How to Prepare for AGI According to Reid Hoffman" ## Speakers A two-person conversation: an **interviewer/host** (Dan, per the outro) and a **guest** (Reid Hoffman, discussing his book *Super Agency*). --- ## Interviewer Delivery Patterns ### Collaborative Framing Before Questions The host consistently builds conceptual scaffolding before asking anything, often quoting the guest's own ideas back: > "one of the core parts that I read in that that I want to pull out is you talk about agency, but then you talk about it as a sense which I read as like a internal kind of like aesthetic experience of agency" - This **quote-then-question** structure appears repeatedly: the catechism discussion, the "compass" question, the "high agency" counterpoint. ### Recurring Hedge Fillers: "like" and "um" Heavy density of "like" and "um" throughout the host's speech, giving a fast, thinking-out-loud cadence: > "Um, which which I think most people when they when they think about agency like it's it's it's all external." - Word repetitions ("which which," "it's it's it's") appear frequently, suggesting extemporaneous delivery rather than scripted questions. ### Pushing for Concreteness A repeated teaching move: the host explicitly asks to move from abstraction to practice: > "I want to make this really concrete though cuz like let's say so we can we can talk about compass what we want. Like how do we fill that out in a person who's like okay maybe I work in tech" - Echoed earlier: "Maybe I'm a product manager, I'm a designer, whatever. How does this change how I think about and orient to AI over the next couple years?" ### Playing Devil's Advocate The host raises anticipated counterarguments to test ideas: > "like a counterargument would be like isn't that a little wishy-washy? ... you got to be first principles thinking on some of this stuff" ### Willingness to Self-Disclose The host shares personal material mid-interview (the OCD tracking app segment), shifting from interview to genuine dialogue — an isolated but notable moment. ### Casual Interjections and Humor Quick quips punctuate the conversation: > "never argue with people at Davos. That's mistake number one" > "Talk about ego." / "Talk about agency." > "every time I thought I was out they pull me back in is Wittgenstein." --- ## Guest (Reid Hoffman) Delivery Patterns ### Hedged, Qualified Claims Hoffman frequently softens assertions with "I think," "you know," "kind of," and "to some degree": > "So um you know part of kind of our fundamental concept of our place in the world..." - This hedging is pervasive enough to count as a stable pattern. ### Concrete Everyday Analogies A signature teaching habit: grounding abstract points in mundane examples (Uber, driving on the highway, GPS systems, Whoop data, hiking trails, birthday sonnets): > "is it a a loss of agency to being driven an Uber or a gain in agency to being driven in Uber?" > "I drove a German one which when I left the path was like return to the path, return to the path" ### Historical References as Evidence He repeatedly reaches for historical analogies (printing press, Socrates, Martin Luther, post-WWII institutions, Einstein and dice) to frame current technology — a stable structural pattern in this conversation. ### "So, for example" Launch Moves He repeatedly pivots into examples with formulaic openers: > "So, for example, think about just driving down the highway." > "So, for example, is it a loss of agency..." > "I'll give a personal example. So, when I first start sat down with ChatGPT..." ### Concession-Agreement Rhythm Hoffman often answers by first agreeing with or extending the host's framing ("Exactly," "Yes," "I completely agree," "Well, I think you know...") before adding his own layer: > "I completely agree and actually one of the things that's interesting about..." > "Exactly. So, it's unclear." ### Quoting Rephrasings of His Own Catechism He self-consciously flags paraphrase: > "the strength to change the things I can, the tolerance to live with the things I can't, and the wisdom to know the difference... I and I paraphrased it." --- ## Shared / Interactional Patterns - **Laughter-free but playful rapport**: Humor runs through both voices (the "Wittgenstein's godfather" exchange, "Or you have a massive ego problem" / "Or both"), showing a conversational, riffing dynamic. - **Interruption-free turn-taking**: Turns are long; the host asks one layered question, the guest responds at length. - **Interviewer echoing**: The host frequently restates the guest's conclusion before building on it ("And that gets to another point that you make in the book..."). --- ## Uncertainty Notes - This transcript appears to be a **looping "best of" edit** — the opening segment repeats almost verbatim near the end (compare [0]–[65] with [705]–[795]), so apparent repetition of phrasing (e.g., the catechism and "throwing yourself under the wheels") partly reflects editing, not only habit. Both occurrences are quoted as they appear. - Transcription artifacts ("Chat GBD," "Vickenstein," "Beijian probabilities," "heruristic") mean exact spoken wording cannot be fully verified from this text. - The closing segment (like/subscribe outro) is a tonal outlier — likely pre-recorded or stylistic — and may not reflect the conversational delivery above. - No timestamps beyond transcript indices were available; delivery features like pacing, tone of voice, and pauses are **not observable** from this text and are not claimed. # Observation Record: "Best of the Pod: How to Prepare for AGI According to Reid Hoffman" **Speakers:** Reid Hoffman (guest, author of *Super Agency*) and Dan Shipper (host, identified via closing ad-read: "Dan Shipper as the captain of the spaceship"). --- ## 1. Historical Analogy: The Printing Press **Hoffman** frames public fear of AI as paralleling reactions to the printing press: - Contemporary worries were that printing would cause "collapse of our trust in human cognition," widespread misinformation, and erosion of the "solidity of our knowledge and society." - Centuries later, we are "deeply indebted" to it: no science, widespread literacy, progress of knowledge, or strong universities without it. - **Qualification preserved:** the transition included "nearly a century of religious war." Hoffman's conclusion is that transitions are challenging but are "an opportunity not just a notion of fear because... the world is what we make it." **Shipper** presses on whether the Reformation-era conflict was preventable — could someone have told Luther to "cool it"? - **Hoffman:** "It's unclear." Luther tried working within the system; it's unclear the Catholic Church would have tolerated gradual reform. Post-WWII global institutions and recorded history may let us navigate this transition "much cleanly and much more compassionately." - **Qualification:** "I don't think there will be zero transition. I think that's pipe dream." ## 2. Theory of Agency **Hoffman's core claims:** - Agency is both **external and internal** — "it's a combination of that internal and external." - New technologies are typically experienced by institutional leaders as reducing agency, triggering fear ("change in agency, therefore reduction and therefore bad"). - **Examples:** Uber (loss of agency if framed as "my hands aren't on the steering wheel"; gain for hundreds of millions who use it to get home safely); smartphones ("billions of people with them because... it's a massive increase in agency"). - **Key line:** approaching an external circumstance as taking your agency means "you're essentially throwing yourself under the wheels"; the alternative is using it to "transform... extend and enhance my agency." - **Super agency** (thesis of the book): when millions gain agency from a technology, "we collectively get super agency both as individuals and as societies." **Shipper's reading and additions:** - Reads agency in the book as a "sense," an "internal kind of like aesthetic experience of agency," connected to **Dennett's intentional stance** (Hoffman endorses this connection, adding "there's an aesthetic stance too"). - Agency is partly "where you put your attention" — not only attention; external things matter. - **Shipper's counterpoint on "high agency":** internal locus of control over things totally out of your control is "a pretty miserable way to live." **Hoffman agrees:** "it definitely doesn't make you more effective." ## 3. Hoffman's Personal Formation - Adopted early in childhood the catechism: "strength to change the things I can, the tolerance to live with the things I can't, and the wisdom to know the difference." He did not grow up Catholic; he "came to love that catechism later." - Attributes this sense partly to **board games**: Avalon Hill games, Starfleet Battles, Dungeons & Dragons, plus chess and Go. - **Reasoning:** games with dice rolls "more closely approximated the kinds of circumstances we encounter in life" — life is not deterministic; chess and Go have "no epistemic uncertainty." - This became his approach to strategy in school, companies, and investing. ## 4. Relinquishing Control **Hoffman:** friendship, romantic relationships, team sports, "team sports within companies" (shared agency), and religious experience all involve "giving yourself over to not being in completely control." **On investing:** - His heuristic: "would I do this investment and walk away and say call me in 5 years" — roughly "98% of good investments" pass this test. - Founders work 80–100+ hours/week; an investor can give ~4 hours. "If they're not capable of carrying the game themselves... you've made a bad investment." - Exception: helping with one specific deal, hire, or strategy element. - **"Iatrogenic investing"** (Shipper's term, Hoffman endorses): investors over-involving themselves and "messing it up" — or, as Shipper adds, "a massive ego problem." ## 5. Uncertainty **Shipper's thesis:** the Western default (Socrates/Plato through Descartes, science's fundamental laws, religion's certain world-to-come) is to try to eliminate uncertainty; grinding against it lowers agency. He does not "denigrate" those outlooks — "quite useful in certain circumstances" — but for technology "that way of thinking... can lead to trouble." **Hoffman largely agrees**, with a self-described "broad brush almost Jackson Pollock" generalization: - People "delude ourselves to over degrees of certainty" (highway driving risk example). - Distinguishes uncertainties: how the world plays out, epistemology, how the system comes together. - Entrepreneurs who believe they're "guaranteed success" are "massively deluding themselves" — though that delusion "maybe sometimes is necessary to work really hard." - Critiques Western philosophy's drive to "human nature is X" (e.g., reducing all self-orientation to selfishness); notes unresolved tension between relativity and quantum mechanics ("God does not play dice"). ## 6. Blueprint vs. Compass (Techno-humanist Compass) **Hoffman:** - Rigid plans "break fast and uselessly"; flexible planning via **compass** — adjusting as terrain and conditions are discovered. - An AI chatbot functions like a "compass" for informational decisions — an "informational GPS." - The **techno-humanist compass**: people treat only the new thing as "technology," but cars, phones, laptops, glasses, books are all technology that acquired a "patina of culture" (Shipper's phrase; Hoffman: "Exactly"). **On first principles vs. compass (Shipper's challenge that compass seems "wishy-washy"):** - Hoffman: the compass "can be amongst a first principle." - **GPS anecdote:** German systems demanded "return to the path"; Japanese systems recalculated ("you want to go a different way. Here's your new set of directions"). "If you say first principle is only being Germanic, then you're going to have problems." - Shipper notes Tolstoy's *War and Peace* describes German strategy the same way; both agree modern systems now recalculate. ## 7. Practical Advice for Workers Facing AI **Hoffman:** - **Iterative development:** start with individuals experimenting. We're in the "cognitive industrial revolution"; no professional can say "I don't use computers" — "AI is just the amplification of that." - Use AI for trivial things *and* "things that you are seriously in earnest about. And you may find that some of them it's not ready yet... some of them it is." - **Personal example:** ChatGPT gave a wrong "business school professor's" answer on how Hoffman would invest in AI, but in the same session produced a due-diligence checklist where "items like four and seven... I would have thought of that in a few days, and it's useful to have it now." - **Jobs:** some (customer service, "trying to get human beings to act like robots") will have "much higher replacement coefficients"; many others — "the human will be replaced by a human with the AI. And we want as many of the humans to be the humans with AI." - If a job goes away, AI can help map skills to adjacent roles (account management, sales, support desk) and help learn/obtain them. - Marketing example: AI as "a co-pilot," not "plug it into the machine and press play." ## 8. Equitable Access **Hoffman:** put the technology in "as many people's hands as possible" — both for fair participation and because unlocking talent from the broadest range benefits everyone. "Equitable access is not just a question of... is it fair... it's also better for all of us." ## 9. Regulation and Private Commons **Hoffman on regulation:** he is "regulatory cautious" rather than **anti-regulation** (correcting Shipper's framing). Reasoning: regulators assume a degree of certainty and "enshrine the past against the future." Start by measuring feared harms, track them, and let innovation reduce them. "Life is risk... by taking some risk we can figure out things like cures for diseases." **On private commons:** - Counters "surveillance capitalism" framing: Google Maps letting others find your house also lets friends visit. - Corporate incentives are "very pro-social" *but* some incentives push lock-in ("you do them only with me") — occasional, specific intervention may be warranted: "I have the right to move that [data] as I like and you must facilitate that." **Shipper's concrete case:** he has OCD (managed, not cured) and built an app in Windsurf combining daily Y-BOCS scores, Whoop data, and video logs to try to predict OCD symptoms from facial, vocal, and physiological data. Whoop's API doesn't give full data access — an example of where portability matters. He predicts the findings will be "thousands of different things... all like weekly [weakly] interacting," and claims "quantified self is... going to happen finally." **Hoffman's response:** agrees this kind of portable data can "create great things" (Whoop, glucose monitors, watches); cites a doctor who can tell from Whoop data "whether or not you're about to get COVID." ## 10. Symbolic vs. Subsymbolic AI **Shipper's framing:** philosophy moved from universal truths (Kant onward: pragmatists, later Wittgenstein, continental philosophy) away from definitional certainty; AI followed — symbolic AI (brittle, computationally explosive) gave way to subsymbolic pattern-matching. His model, "open to revision": "symbols are important but they are important when they arise out of a subsymbolic architecture." He adds the Wittgensteinian "therapeutic" point — a philosophy that lets you stop philosophizing — though mostly "rhetorical," since he doesn't want to "cast aspersions on symbols." **Hoffman:** his undergrad major was **symbolic systems**; he was in the early **connectionism** movement holding that symbols-only theories "radically underperform." References Wittgenstein on rule-following. **Projection (tentative):** over the next 10 years, a "play between probabilistic models and symbols" beyond current transformers. Example (from the past, "not saying this is the right one"): symbols with **Bayesian probabilities** attached ("we're having philosophical insight... probability 90%"). **On the o1 paradigm (Shipper's question):** Hoffman — not "unusually" different from other transformers, but chains of thought with a fitness function, and "that fitness function itself might have that probabilistic characteristic." **On Wittgenstein:** Hoffman notes even Wittgenstein returned to philosophy after claiming to have solved it — "which is the reason why he... died as a philosophy professor." Shipper: "Every time I thought I was out they pull me back in is Wittgenstein." ## 11. Science vs. Engineering; Reasoning Modes **Shipper:** post-Newton as "the era of science"; AI is "turning science problems into engineering problems" — more pragmatic, less insistent on explicit causal explanations, more tolerant of "lots and lots of little correlations." **Hoffman adds:** a shift from **deduction** toward **induction and abduction**. Abduction defined as "the best theory... theory that models the evidence," derived from priors/creative process rather than the data alone (induction = modeling the curve from data points). "When we're good thinking beings, we apply all of them." **Shipper on moral theory:** no hole-free moral theory exists; taking effective altruism "too far" leads to "SBF" — an explicitly stated caveat that "effective altruism is a great idea" beforehand. He argues for a "diverse partially explicit but... mostly implicit set of moral theories" — "fundamentally sort of philosophically unsatisfying" but "kind of beautiful." **Hoffman:** theories of world, morality, selves should be treated as **dynamic and updated** — not just from new data but "the way we learn from each other" — leading to "best judgment" circumstances. "By the way, that's how science progresses too." ## 12. Purpose of the Book (Hoffman's Stated Audiences) 1. **The AI-curious and skeptics:** "lenses" for seeing why AI is "humanism and humanity positive." 2. **Technologists:** human agency as "almost like a design principle" — a fitness function for design and deployment decisions. **Design examples given:** - ChatGPT's iterative deployment (GPT-3.5 existed ~a year but ChatGPT made it accessible — affordances matter). - Caution against defaulting to **human replacement** (e.g., customer service, citing Sierra); prioritize the **co-pilot** model — "amplification intelligence" (per his book *Impromptu*). ## 13. Lowering the Cost of Intelligence Services (Shipper) - His ghostwriter example: he can afford a ghostwriter for YouTube copy; beginners can use ChatGPT, Claude, or his company's internal incubation "Spiral" — reducing the leverage gap between wealthy and starting-out creators. **Hoffman's response:** AI "raises the bar" for those with less *and* amplifies those who already have advantages (a ghostwriter using ChatGPT). "Amplifying across the board is actually a good thing." --- ## Unresolved Tensions Preserved - Whether historical transition conflicts (Reformation wars) were preventable — Hoffman: "it's unclear." - High agency as ideal vs. the value of relinquishing control — both speakers hold both, without a resolution rule. - Shipper's "anti-regulation" characterization of the book vs. Hoffman's self-correction to "regulatory cautious." - Hoffman's 10-year projection on probabilistic-symbolic integration is explicitly tentative ("not saying this is the right one"). - Shipper's claim that quantified self "is going to happen finally" is his prediction, not a settled shared conclusion. # Speaker Attribution Record — Video `01a0f15a-db8f-72a1-84a2-3b1ca501c090` ## Video - **Title:** Best of the Pod: How to Prepare for AGI According to Reid Hoffman ## Speakers & Roles ### Primary Guest - **Reid Hoffman** — Guest, interviewee. Author of the discussed book *Super Agency*; also references his earlier book *The Startup of You* and *Impromptu*. Identifies himself as an investor and former undergraduate in **symbolic systems** (early connectionism). Referenced as "Reed" in the host's greeting. ### Host / Interviewer - **Dan Shipper** — Host/interviewer. Identified by name only in the closing promotional segment ("Dan Shipper as the captain of the spaceship"). States he runs a media company, has a ghostwriter, has OCD, and built a personal tracking app using **Windsurf**. Conducts the interview and asks the questions throughout. ### Promotional Narrator / Sponsor Voice - **Sponsor read (unnamed voice)** — Reads the mid-roll ad for **Adio**, an "AI native CRM," including a discount code (`adio.com/every`, 15% off first year). Names Union Square Ventures, FlatFile, and Modal as customers. - **Closing promotional voice (unnamed, likely AI-generated or show voice)** — Delivers the "smash that like button and subscribe to AI and I" outro, referencing "chat GPT," and addresses "Dan" directly. **Flagged as promotional.** ## Provenance Notes ### Referenced Works & Ideas - **Reid Hoffman, *Super Agency*** — the interview's central subject (upcoming/recently released book). - **Hoffman, *The Startup of You*** and ***Impromptu*** — cited by Hoffman as his prior books. - **Dan Dennett's "intentional stance"** — cited by Hoffman. - Historical/philosophical references by both speakers: Socrates, printing press, Martin Luther/Reformation, Sun Tzu, Tolstoy (*War and Peace*), Wittgenstein (early/late), Kant, Descartes, Einstein/quantum mechanics, Alcoholics Anonymous serenity prayer, "surveillance capitalism" critique, effective altruism / SBF. ### Third-Party Products / Claims Referenced - **AI/tech:** ChatGPT, GPT-4, GPT-4o1, o3 (upcoming per host), Claude, Google "Titans" (released "today" per host), Windsurf (Cursor competitor), Sierra (customer service AI), Whoop (health data/API), Y-BOCS OCD assessment. - **Sam Altman** — quoted/paraphrased by host claiming AGI is essentially achieved (third-party claim, unverified in source). - **Hoffman's anecdotal claims:** Davos smartphone debates; German vs. Japanese car GPS behavior; his doctor detecting COVID onset from Whoop data. ### Promotional Segments (Flagged) - **Mid-roll sponsor read** for Adio (CRM product with discount code). - **Closing subscribe/like promo** for the show "AI and I," naming Dan Shipper. - Host's positive endorsement of *Super Agency* ("I read it. I actually really liked it... professionally I have to say that") — noted as a potential conflict-of-interest disclosure by the host himself. ### Uncertainty / Availability Limits - Transcript lacks speaker name labels; attribution of host as **Dan Shipper** rests on the outro naming him and the host's self-described media-company role. - No timestamps, publication date, channel metadata, or comment availability data were supplied. I think there's many times when you're encountering an external circumstance. If you approach it as this is taking my agency, you're essentially throwing yourself under the wheels, right? Whereas you go, oh, here's how I can use this to transform my agency, to extend and enhance my agency in various ways, then it becomes much better. \>\> One of the core parts that I read in that is you talk about agency as a sense internal kind of like aesthetic experience of agency. I think one of the things I picked up from fairly early in my childhood was realizing that old catechism that's pretty good which is, you know, the strength to change the things I can, the tolerance to live the things I can't, and the wisdom to know the difference. I came to love that catechism later because I had early got to that sense of that's how you should navigate the world. \>\> Wow. How'd you do that? \>\> Maybe it's like playing a lot of board games. I mean, it's kind of like like it's kind of a sense of, hey, these things are under your control and these things are not under your control. \>\> You have a couple of like specific ways of thinking about that that I think are really interesting. So, I want to make this really concrete though. Maybe I'm a product manager, I'm a designer, whatever. How does this change how I think about and orient to AI over the next couple years? Arm me with some more practical, concrete ways to instantiate this compass in how I approach my life. Reed, welcome to the show. \>\> It's great to be here and great to be doing this in person. \>\> Yeah, I love that. Um, it's just it makes it a much different experience just to like be face to face. We're actually very close to each other. \>\> Yes, exactly. Well, and actually even philosophically. \>\> Yeah, definitely. Um, so I'm excited to have you for a number of reasons, but one of the big ones is you have a new book uh coming out by the time this is out. It may be out. It's called Super Agency. Um, I read it. I actually really liked it. Um, it's part of my uh like professionally I have to say that, but also I really \>\> but but it happens to be true. \>\> It happens to be true. And the reason I like it is um I think it's it you're you're writing it in response to um the I you know there's a prevalent fear of AI and maybe AGI as socially. Um, and I think you're uh you're examining this moment and saying if we want to understand what to do in this moment and how things might play out. A good way to do that is to look at history and look at other um historical technology changes to see like um how we reacted and whether we were wrong or not. And you go much further than that in the book, but I want to start there because I love I love that history angle because I think we forget how many of the things that we're familiar with now uh used to be really really scary. I wonder if you could go into some of those uh so we can understand it a little more. \>\> Well, um it goes all the way back to the written word with Socrates. \>\> Um although there are some different scholarly interpretations of the Socratic remarks or the uh challenges. So why don't we start with the printing press, \>\> right? \>\> Um and you know when the printing press was introduced a lot of the public dialogue was very similar to the dialogue we have around artificial intelligence. Um this will lead to the collapse of um our trust in human cognition. Um it'll lead to uh widespread misinformation. It will lead to um the collapse of the solidity of our knowledge and society and what we're doing. And so there and who knows what people are going to do with this technology that could really erode things. And you know on one hand you know here um many centuries later we are obviously deeply indebted to the printing press. We can't have science without printing press because you can't get to that spread of information. You can't have uh widespread literacy and education. You can't have the progress of knowledge uh middle class institutions universities etc. All of which well universities existed before the printing press but getting to the the the many universities of vigorous and strength required the print printing press. Now so all of that shows that it's great despite all the fear. Now that being said the other thing the the thing to track with the printing press is u there was nearly a century of religious war. So we as human beings uh when we get these new technologies the transition periods of what we get in can be very challenging. But that's is you know to some degree that's a that's an opportunity not just a notion of fear because it's like the world is what we make it. So let's make this transition much better than the earlier transitions. \>\> Do you think that cuz cuz one of the one of the you know overriding points in your book is um these kinds of transitions we have some amount of control over them but we don't have total control. So, and and we can get more into like all the nuances of what that means, but I'm curious about um do you think we can prevent, you know, like let's say uh we went back to the printing press days. Uh you know, is there something we could have said to Martin Luther to be like, "Hey, like this schism thing is and the Reformation, like it's great to have a personal relationship with God, but like maybe cool it a little bit on some of this stuff because it's going to lead to a lot of war." Um or is that sort of inevitable? Well, I think it depends on where your society and culture is at the time. Um, it's unclear that even if Martin Luther had said, "Let's do this gradually and have a set of discussions that the reigning Catholic Church would have tolerated that and allowed it." \>\> I think he did like I think he was trying to go within the system for a long time before he did the the door pounding thing. \>\> Exactly. So, it's unclear. But one of the things we hope in having built, you know, great global institutions post World War II, having learned from what the the massive amount of tragedy is when you get in that kind of amount of conflict. \>\> Yeah. \>\> That and by the way, of course, we have the uh we have various technologies, not the least of which is printed books to remind us of our histories and to learn it that maybe we can do this one much better. Yeah. \>\> And that's part of the reason why in you know as you know from having read super agency it's part of the reason for for doing the book for doing podcast and other things is say hey if we actually steer the right way we can minimize the transition difficulties. I don't think there will be zero transition. I think that's pipe dream and and moment we should have that expectation. But we can actually get through the transition much cleanly and much more compassionately and humanely than we have done in earlier transitions and get to the amazing benefits that that is always the other side of these m these massive new technology leaps. \>\> That makes sense. I think one of the um one of the places to go from there is it's it's the book to me is not just a history book. It's also um a little it's a book about psychology in a certain way because um you have a theory of why we are typically very afraid of new technology that runs through all new techn technological advancements. I wonder if you could tell us a little bit about that. \>\> So um you know part of kind of our fundamental concept of our place in the world and our you know dignity, our meaning etc comes down to a notion of agency. Um, and it's both agency as individuals and agency as groups and agency as societies. And so most often, and you know, back to like the earlier referred printing press, what people experience the new technology as is reducing the agency um, for you know, kind of uh, you know, key people who are leaders in society, you know, heads of institution, participants in institutions, the institutions themselves. and and so they they react because they go, "Oh, this is going to be a change in agency, therefore reduction and therefore bad, therefore destructive of society." And the actual path that happens is yes, agency changes. So some things that you had agency in before you no longer have agency in. But when you begin to look at agency, it's not just a set of external factors, it's internal factors. It's how do you approach it? So, for example, like let's let's use a very modern example that people can think about. So, is it a a loss of agency to being driven an Uber or a gain in agency to being driven in Uber? And obviously, if you're like, "Oh my god, I don't my hands aren't on the steering wheel and who knows what this random human being is doing." And and D then you're like, "Oh my god, it's enormous loss of agency." And yet, of course, you know, hundreds of millions of people are doing it because they realize it's a gain of agency. It's I I can I can I can not have a car and get somewhere. I can I've gone, "Oo, I drunk too much. I'm going to get home this way much more safely, etc., etc., etc." And so, it's a question of how we approach it and how we experience and how we choose our own agency. That doesn't mean it's completely separate from external effects, but it's a combination of that internal and external where it's really appropriate. And then what happens as we like we first encounter technology like like I remember arguing at Davos with people about um our smartphones these kind of humanity reducing you know kind of um like almost like cybernetic control of human beings you know and \>\> never argue with people at Davos. That's mistake number one is uh you're making me think about a princess bride, right? Your first mistake. Uh and uh and then uh obviously part of the reason why we have billions of people with them is because actually in fact it's a massive increase in agency. And matter of fact, it quickly gets to, you know, kind of the thesis of the book of super agency is what happens when millions of people get all elevated in their agency with a new technology and then all of a sudden we collectively get uh super agency both as individuals and as societies. \>\> Yeah. I think um one of the one of the core parts that that I read in that that I want to pull out is you talk about agency, but then you talk about it as a sense which I read as like a internal kind of like aesthetic experience of agency. Um which I think most people when they when they think about agency like it's it's it's all external. Um but it's quite clear to me that that that aesthetic component is a huge part of it. And um and and I think a lot of the the questions about human agency um they reflect almost like a lack of faith in our ability to change and adapt and then also lack of understanding of like how that aesthetic sense arises and how it you can bring it to any experience um to some degree like the the history of literature is like about people in like pretty dire circumstances feeling like they have some agency. Um, which is not to say that external conditions don't matter. Um, but a lot of it is internally driven. I'm curious like how you got there, like what influenced that? How how did you start to think about this? \>\> Well, some of it may be uh philosophical background and and the highlighting the aesthetic is also very interesting. I think maybe when I was writing I was also thinking of like Dan Dennett's intentional stance like it's a stance that we have towards the world like a mental stance a worldview and obviously there's an aesthetic stance too. So I think that's a great uh highlight and you know I think part of it is that you know it's like using the Uber example and there's and using the I the the smartphone example um I think that there's many times where um when you're encountering an external circumstance including of course technologies if you approach it as this is taking my my agency then you are um you're essentially throwing yourself under the wheels Right? Whereas you go, "Oh, here's how I can use this to transform my agency, to extend and enhance my agency in various ways." Then you then it becomes uh much better. So, for example, think about just driving down the highway. You go, "Hey, I have I'm in this car. I have an ability to slow down, speed up, drive, etc." Well, there's a lot of other cars on the road, too. If you say, "Oh my god, my agency is is taken away because these other cars are slowing down or are potential hazards, etc., etc." Well, then you're never going to get on the road, right? \>\> Yeah. And and that I think that points to like um agency is to some degree it's a it's a way of looking at the world and where you put your attention. So like if you push your attention to oh my god, all these other cards, of course, that that sense of agency is going to go down. If you push your attention somewhere else, that sense of agency is going to go up. And again, it's not it's not only your attention, like there are external things, but like to some degree, there's there's some sense of control. And I'm curious um for you, how how do you notice your personal sense of agency fluctuating dayto-day and in your life? Like have there been periods, have you always felt um this connection with a sense of agency? Have there been periods where you haven't? And like how has that played out for you? \>\> Great question. You know, it's interesting because I think one of the things I picked up from fairly early in my childhood was realizing that kind of that that that old catechism that's pretty good, which is, you know, the something like the strength to change the things I can, the tolerance to live with the things I can't, and the wisdom to know the difference. \>\> Yeah. It's like the alcoholics anom anonymous prayer thing. \>\> Yes. Exactly. And I think it comes from a uh a Christian Catholic catechism. And and I and I paraphrased it. \>\> Did you grow up Catholic? \>\> No. No. No. No. Just uh it wasn't so much that as much as I'd adopted that I I came to love that catechism later because I had had early got to that sense of that's how you should navigate the world. \>\> Wow. How'd you do that? \>\> I think it just maybe it's like playing a lot of board games. I mean it's kind of like like it's kind of a sense of hey these things are under your control and you can affect them and then you can affect that outcome and these things are not under your control and overly like tearing your hair out about the things that are out of your control is not helpful to you or to anyone else. Right? So you go you know like um uh the fact that there's suffering in the world. If you go, "Oh my god, there's suffering in the world." Then of course you're going to get crushed. There's going to be suffering in the world. You should of course always feel for people suffering. But the fact that there is like children dying around the world today, it's really really sad. We should try to do things about it, but we're not going to stop all of it today. It's an ongoing process. And so, so there's things out of your control, and then there's things in your control. And of course, that's where your ability to navigate it. Um and so it's kind of like whether it's um you know maybe it's kind of simple like who you're friends with when you're in school and and you go oh you know like you know do you have friends that you treasure and then that's really great and maybe there's other people that you wanted to be friends with weren't as interested in you that's fine too right but kind of navigating that within kind of child circumstances I think where I where I started and then from there I think that the um that became kind of how I approached each new challenge that I was getting and you know part of how like I think I got a sense of good strategy in life and strategy for me, strategy for what I did in school, strategy for what I did with companies, strategy for what I do with investing all came from this kind of uh like figure out what the nature of the game is and what are the things that were within your ability. ability to change and then accepting the things that you can't while uh like changing some really interesting things. \>\> And so and it sounds like um that came from board games. What board games were you talking about? \>\> Well, so um a whole set um some uh and there's less board games, Dungeons and Dragons. So I was I was doing that uh which there seems to have been some resurgent on which is cool. Uh but also what probably most people don't track these days, these Avalon Hill board games. I did a bunch of um and then a variety of others and and one of the things about it being multiple is that you're learning all of them. Um and actually one of the things that people frequently said and I did play chess and I did play Go. Um I wasn't as attracted to those games because part of the thing with when you're playing like the Avalon Hill board games or also like Starfleet Battles was another one I did. um by having some randomness with dice rolls, it actually more closely approximated the kinds of circumstances we encounter in life because um life is not like chess. Life is not like go. It is not deterministic that way. There's uncertainty variables that you have to you have to play into and and and epistemic uncertainty sometimes that you have to play into and and both go and chess have no epistemic uncertainty. And so uh uh adopting your strategies to those were really important. I think that's those are those are the initial mindsets. \>\> That makes sense. I think um there's this there's a meme in tech right now about being high agency like high agency people which I think is great. It's good to be high agency, but I think um we tend to think of agency as always good. \>\> And generally if you if you want to be like generally I think it's good to have agency, but for example in in the in the um uh in the example you gave of there's suffering, right? If you think of yourself as high agency and have a high locus of control, internal locus of control for things that are totally out of your control, um it's actually a pretty miserable way to live. \>\> Yes. Um, and it definitely doesn't make you more effective. And so there's a there like I think what you're saying is there's a certain range of things within which you want to be high agency and have an internal locus of control. And then there's also a whole set of other um experiences in life that are important that are about completely giving up control and recognizing um control. And those are some of the most recognizing your lack of control. And those are some of the most meaningful experiences that people have like transcendent experiences. Um curious what what that sparks in you. \>\> Well, I mean I think the obvious ones are in friendship, romantic relationships and other things is part of what you're doing is um is essentially uh giving yourself over to not being in completely control of how a relationship's playing. And those are obviously some of the places where as we learn and become, you know, kind of wiser, more compassionate, more evolved people. Um that's actually in fact, I think one of the the really central line things. It's also, by the way, sometimes um people encounter that playing in uh team sports of various sorts. It's there's team sports themselves. There's also of course team sports within companies in terms of how a company's operating the shared controls the shared agency um becomes I think uh you know really key and obviously some people find that within a religious experience too. \>\> Yeah. How has that um worked for you as an investor obviously because you're you know you're investing in companies you care about the outcome deeply but also uh you know you don't want to be too in control in a lot of ways. Uh there's a lot of trade-offs to control there. How's that worked for you? \>\> Well, I think one of the things that um so start with something kind of enormously simple, pragmatic, heristic. Um one of the things questions that I ask myself on an investment is would I do this investment and walk away and say call me in 5 years \>\> because um almost like call it 98% of good investments you will accurately have that sense. Not to say you wouldn't try to help and work with them because well like part of it is you think as an investor you know I can spend maybe if I'm spending a lot of time on investment I might be able to spend four hours a week on the investment because as an investor I have a portfolio of investments I've got other things I'm doing and so the person I'm investing in uh she or he um you know the the CEO the founders um you know how they may all may all play out they're in this presumably 80 plus hours a week, maybe 100 plus hours a week. And so if they're not capable of carrying the game themselves, it's almost for sure you've made a bad investment. Now, sometimes the exception is where you go, oh, what I really need to do is help them get this one deal or help them hire this one person or help them with this one strategy element and then everything else will be fine. Then that's okay and I will sometimes do that investment. But one of the mistakes that investors frequently make is they go well, you know, they'll they'll tell themselves that they're more important than they are in in the kind of the in the the judgment and then they'll get too much in involved in it and they'll actually be messing it up. \>\> Yeah. \>\> Because iatrogenic investing. \>\> Yes. Because truly if you think your 4 hours is more important than the person's 80 to 100 hours like like you probably invested in the wrong person. \>\> Yeah. Yeah. Or you have a massive ego problem. \>\> Yes. Yes. Or both. \>\> Probably. \>\> This episode is brought to you by Adio, the AI native CRM built for the next era of companies. With Adio, setup takes minutes. Connect your email and calendar and it instantly builds a CRM that mirrors your business with every contact enriched and organized from the start. From there, Adio's AI goes to work. It gives you real-time intelligence during calls. It prospects leads with research agents and it automates your team's most complex workflows. Industry leaders like Union Square Ventures, FlatFile, and Modal are already building the future of customer relationships on Adio. Go to adio.com/every and get 15% off your first year. That's atio.com/every. And now back to the show. Um, yeah. So, I I think one of the things that um that that came to me as like as part of this book is your is your tracking agency as this like psychological experience, an intentional stance, an aesthetic stance. Um and that one of the things that um seems to affect our sense of agency is how we approach uncertainty. And the more we kind of grind against uncertainty to try to like um eliminate it, uh the the lower our agency is because you can't eliminate it. Um and the more you like you work with uncertainty, um and have that sort of stance of like surfing through it, um the better things go. And that's the appropriate stance for like dealing with new technologies. And the thing that of course for me like the thing that I'm thinking about is well I think our default stance in the west to uncertainty is to try to eliminate it. And you can like read that in the history of philosophy um which is like um you know uh Socrates and Plato to Daycart trying to make knowledge that is like clear and explicit is epistem um and it's the same thing in science right like you want to boil things down into fundamental laws so you can predict things in advance um laws that are explicit religion I think is also in a lot of ways about dealing with uncertainty by trying to eliminate it um maybe uh the world of the senses is totally uncertain, but there is a world to come that is like totally certain and is definitely going to happen and and will be amazing. Um, and and not to denigrate any of those things because I think those outlooks are actually quite useful in certain circumstances. Um, but in particular, um, uh, when it comes to technology, um, that way of thinking about things can lead to trouble. I'm curious if you have the same reading or or how you would respond to that. Well, I think it's definitely the case that we and I think there's the the the generalization uh which you know is kind of a very broad brush almost Jackson Pollock you know you know quick quick quick uh description of the stuff but I do think that the notion that we as human beings try to um delude ourselves to over degrees of certainty like we don't for example people realize every time we drive somewhere on the highway, we are taking a certain amount of risk. We're taking a risk about our own competence, we're taking a risk on our vehicle, we're taking a risk on weather conditions, we're taking a risk on other folks. And of course, once again, just like agency, if you to dwell on all that, maybe you'd never get into a car, but then you would never go anywhere, \>\> right? And never be able to benefit from, you know, significant portions of modern society. And so um and so part of it is to think about how do we live in a heruristics of managing uncertainty and managing various forms of uncertainty like there's a form of uncertainty about well how does the world actually play out. There's uncertainty about our epistemology and how we know the world and then there's uncertainty about how the system comes together. And um you know it's one of the things that's really central to entrepreneurship because any entrepreneur who persuades themselves that they're guaranteed success in what they're doing is massively deliluding themselves, right? Which maybe sometimes is necessary to work really hard in order to do something. But you you know there's always like uh encountering and navigating a world of risk is actually in fact uh very important and and so I think no I think part of and actually I think that's it's a true line of where frequently philosophy um you know kind of like the western philosophy actually we should probably isolate this too \>\> um you know frequently is overly trying to get to like human nature is X like you know frequently they'll say human nature is selfishness and you're like well there's a lot of different ways in which which human beings um are self-oriented but there's a diversity of them and sometimes you say well when this person's really committed to their family or really committed to their society well that's just the way they're expressing their selfishness well but that's not exactly what you mean by the word selfish. So there's there's a whole bunch of different self-orientation and that that spread is actually really important and the same thing is true for understanding uncertainty well which of course is part of the reason why we're still struggling to resolve um you know uh Einstein's theories of relativity and quantum mechanics and you know Einstein himself was quantum mechanics can't be right because God does not play dice with the universe. \>\> Yeah. I mean, I think that I think that's a really another example of like um how aesthetic sensibilities guide what we see and guide um our decision-m and like supposedly like really rational truth oriented endeavors. And that's why I think I'm sort of keen on this aesthetic sense because I think when we talk about dealing with uncertainty. Um there's a way in which like you can try to eliminate it. You can try to manage it which is sort of still like we'd rather we didn't we we weren't uncertain. Um and then there's a way of like sort of playing with uncertainty and like you use a phrase like bloomers like sort of using that to flourish which I think um in a lot of ways like that is uh that is a way to look at life. Um, and I think that perspective, going back to the aesthetic sense, is often embodied in entrepreneurs and artists who are often guided by by an aesthetic sense. Um, and I think, um, those two outlooks where you're like, yeah, I'm I want to like face the blank page or I want to like, um, change how people see the world, how the world works, um, even though I know it's like highly uncertain is seems to me um, somewhat core to the to to working with technology. I completely agree and actually one of the things that's interesting about the the arc that you're highlighting in your call it philosophical aesthetic reading of super agency is um the actual ties between my first book the startup of you and this because the startup view in part was trying to give advice you give entrepreneurs but to individuals for their lives and part of it was like multiple of the chapters are navigating environments of uncertainty because that's what entrepreneor reneurs must do and must do it intelligently \>\> and preserve their sense of agency while doing it. \>\> Exactly. And so it's an interesting I hadn't quite realized between book one and book six that there was kind of an ar the kind of an arc from it. And I do think that part of the thing in surfing is a very good metaphor is to say how do you embrace uncertainty in a way that you're taking agency and that you are leveraging it to having a better life to working better to making decisions better uh to creating better right in terms of the the the artist side and and well many things but artists especially and Uh, and then leveraging that to a feature, not a bug. \>\> Yeah, totally. I think um the the the the wave metaphor is is really is really interesting to me because you can like try to like stand in front of a wave and stop it um or you can surf it. And the surfing like you're you're working with a dynamic context. So, it's not like you're um uh you're not totally in control, but like you are in control within that sphere, which is sort of what you're what you're going for. um which which I think go gets to your point about um the way our stance toward uncertainty is what um what guides whether or not we feel agency and you have a couple of like specific ways of thinking about that that I think are really interesting. So like um one of the um one of the dichotoies you you bring up is like some people try to make like a real like a blueprint or like a like a plan or like a specific theory about the future. Um and then um uh but you use this the metaphor of um a compass or um or you say cognitive GPS um or another another um term which I don't think is in the book but I think is really related is wayinding. So, it's like um you have a you have a sense of the direction you're going, but you're pretty open to the details. I'd love for you to talk about that. \>\> So, um Exactly. Right. And part of the thing whether or not almost everything if you're making too concrete of a plan and like your earlier, you know, we were kind of talking a little bit about Sunzu and no battle plan survives contact with the enemy. \>\> Um plans that are rigid break fast and uselessly. Yeah. \>\> And so you can make flexible plans, but one of the most easy ways to make a flexible plan is to have, you know, kind of a compass. And the compass might be like, for example, you're like, well, I need to hike from here to here. And there might be multiple trails and sometimes one of them washed out and everything else. Well, you can get from there to there with like and maybe there's no trails if you have compass, map, information as a way of doing it. and you're you're adjusting your plan as you go and as you discover what the terrain and the conditions look like. And it's a good metaphor for also a wide variety of things because as you think about for example well we're making decisions about like what jobs we might take, what careers we might take, where we might go on a holiday, where we um what we might be doing you know Friday night. All of those areformational space questions and again a compass is a good way to do that. And so if you think about what is an AI chatbot helping you with, it's almost like it's a compass helping you with these informationational decisions. Henceformational GPS. And it's resilient because if you suddenly discover something really changes, the world isn't as you expected. The world changed. You change. Like you're like, "Well, I thought I was in mood in the mood for watching a movie, but I'd actually rather just like to have a quiet conversation with my friend." Well, oh, hey, there's a cafe down the street. Let's go. That we discovered that from, you know, uh the the compass. And part of um you know what in writing the book, as you know, we described as the technohumist compass. And the reason is too often when people think about technology, they think about the thing that is new and therefore not really fully adjusted into us. So they go, "Oh, AI, that's technology." Whereas our cars aren't technology, our phones aren't technology, our our laptops aren't technology, our glasses aren't technology. It's like \>\> our books aren't technology. \>\> They're all technology. \>\> Yeah. \>\> And what happens is we uh we we get familiar enough with them that we naturally um uh include them in our existence, include them in our agency. \>\> They build up like a patina of culture. \>\> Exactly. \>\> In our psychology. Yeah. And that's what we want to be doing with this in order to embrace it. And that's why giving metaphors and it being techno but also humanist in the compass. One of the things that the compass makes me think of is um like a counterargument would be like isn't that a little wishy-washy? Like one of the big um uh one of the other big memes or uh viewpoints in tech is like you got to be first principles thinking on some of this stuff. you got to go back to first principle and and and so you're not going to like just change your mind based on like the latest like where the wind is blowing right so how do you how do you square those two \>\> well to some degree I think a technohhminist compass can be amongst a first principle right because the question of principles are what are the set of kind of uh navigational truths that you're using to make decisions to uh incorporate information and change belief states, action states, etc. And a and using a compass as a means of a metaphor versus for example a um you know saying well we only can can be in one route. I think the first principles thing is a very good thing that I myself and I think you as well embrace from a viewpoint of clarity of thinking about things. But like like as a metaphor um when the earliest uh GPS navigation systems for cars played out, I had the entertaining observation between the German and the Japanese systems. I drove a German one which when I left the path was like return to the path, return to the path, return to the path. And the Japanese one was like, "Oh, you want to go a different way. Here's your new set of directions for for what you're doing." And I think the question is is if you say first principle is only being Germanic, then you're going to have problems. \>\> Yeah. Yeah. Yeah. Yeah. Uh it's so funny cuz I was telling you before we started recording that I' I was reading War in Peace over over Christmas and uh that's literally exactly how Toltoyh talks about how German generals do strategy. So I'm glad it made it all the way into their GPS. all the way into their modern GPS. Although I think they've now since learned some from the Japanese ones and they're both like, "Oh, when you didn't turn there, maybe you had a reason for not turning there. We're going to recalculate a new path." \>\> Um, I want to bring this like I want to make this really concrete though cuz like let's say so we can we can talk about compass what we want. Like how do we fill that out in for a person who's like okay maybe I work in tech, maybe I'm an engineer, maybe I'm a product manager, I'm a designer, whatever, and I'm seeing all these changes happening. um how does this change how I think about and orient to AI and think about okay over the next couple years 03 is coming out soon like Sam Alman's saying like hey like we pretty much have got AGI um Google just released something today like called Titans that's that's really cool and seems like a really interesting step in that direction um so give me arm me with some more like practical concrete ways to like instantiate this compass in how I approach my life \>\> so you know another uh concept in the book is iterative development which is part of actually how we get broad-based um networks of inclusion what we also refer to as consent to the govern by customers by you know kind of uh theorists and commentators by government etc uh by press the whole way of feeding back into the system and it starts with individuals experimenting with it right like engaging and I think one of the reasons why when You think that we're in the cognitive industrial revolution and you think oh there's a new set of tools by which like for example I can't be a professional and say I don't use computers I don't use smartphones every professional job requires that kind of information uh connectivity processing analysis consumption generation etc. AI is just the amplification of that. And so every single profession is going to require how am I deploying AI? How am I engaging with multiple agents in order to do my work? Um and so and you say, well, how is that exactly? Do I read a book right now? It's like no, no, the best way is just start engaging with it, engaging with it in some seriousness. So it's fine, for example, to go to chat GBD and say, give me a sonnet for my friend's birthday or my kids, you know, birthday. or whatever, that's great. You know, what ingredients can I have that I can make into a meal for my lunch? Great. Yeah, \>\> do that. But also use it for things that you are seriously in earnest about. And you may find that some of them it's not ready yet, \>\> but you'll find that some of them it is. I'll give a personal example. So, when I first start sat down with uh Chad GBT, I asked it uh how would Reed Hoffman make money investing in artificial intelligence? and he gave me the uh the business school professor's analysis that was completely wrong for venture capital. It sounded smart like you identify large total addressable markets, you understand which products and services would actually \>\> what what a relief. \>\> Yes. And so I was like nope I I'm still I I still have a very unique skill set for doing this relative to these things. But on the other hand, if I then said, "Oh, it's useless for investing," then I wouldn't have have noted the things that it's good for. For example, same session, I sat down and I put in a business plan and I said, "What how should I do due diligence on this business plan?" And you know, the list of things it came back with were not necessarily none of them were a surprise, but items like four and seven, I would have gone, "Oh yeah, I would have thought of that in a few days, and it's useful to have it now. for figuring what to do. And so, so engaging with it is really important because we will see a lot of jobs transform. And the way they'll transform some jobs, customer service, other places where you're essentially trying to get human beings to act like robots, there will be much higher replacement coefficients. But a lot of jobs like the human will be replaced by a human with the AI. And we want as many of the humans to be the humans with AI. So be learning it and adjusting with it. \>\> What is so then like let's let's go deeper into that. So what is an agency preserving way to approach the question which is what happens if my job goes away because of AI. \>\> So first is I don't think that well I can I can I'll do the job goes away too but I think a lot of jobs won't go away. I think they'll transform. And so the question is are you adapting and transforming with them you know at sufficient speed and in advance which is one of the reasons to start playing with AI now. Um and AI by the way can help you with that right because AI can go oh well um here is the new way that we can be accelerating how we understand what a good marketing message would be how we would test it how we would generate options for it. how would you think about what that goes and AI already today can help with a bunch of different marketing tasks and doesn't mean it will do it decisively and say oh we just plug it into the machine and we press play and look that's our marketing because um like I've used GBD4 and GPD01 for this and it's really helpful but it's a co-pilot right for these things so um now in a replacement circumstance like say for example customer service. So you get, you know, companies like Sierra and others who are um kind of designing a how do you have the a the AI agent be the first customer service agent and that may reduce a lot of different customer service jobs and you say, "Oh, my customer service job went away." Well, AI can still be helpful to go, okay, here's what my skill set is. Here's what I think I'm capable of learning. What job should I now go look for? if there's now far fewer customer service jobs. And you might go, well, there's these things in account management, there's these things in sales, um there's these things in uh support desk, there's these things in and that kind of thing may actually in fact go okay and then how do I how do I learn those jobs? How do I get those jobs? Um how do I do those jobs? AI can actually help with all of that. \>\> Yeah. And that gets to another point that you make in the book which is that um one of the really important um agency preserving approaches to new technology is the idea of equitable access. Um can you talk about that? \>\> So one of the things that's uh really key I think for kind of answering this question about look how do we bring as many people along in society? How do we make sure it's not just beneficial for rich individuals, rich uh companies, but as broadly as possible is to have it in as many people's hands as possible. And when you do that, um not only do does it obviously create a better chance of fair and just participation in the new jobs, the new uh career paths, etc. But it also makes society a whole lot better because when you get uh talent from as broad range as possible being unlocked to be to do work to be creative to be to create maximum benefit all of the rest of us in society benefit from that too. \>\> Yeah. \>\> And so equitable access is not just a question of kind of like well is it fair for everybody which it is it's important to have but also that it's also better for all of us. \>\> Yeah. Another another uh extension of that that you talk about that I I feel really uh tied to or connected to is this idea of um uh like generally I'd say the book is like anti-regulation in general. Um competition will regulate itself that that kind of stuff. Um but one of the places where um you seem a little bit more open to the idea of regulation in some form is um uh the idea of um private commons. the way that we treat our data in a world with AI maybe should shift in certain ways. Um so like the the personal data that I've accumulated let's say in Facebook and Google and all that kind of stuff um becomes tremendously more valuable to me as an individual um when uh I have I have GPT4 to go through it and be like here are some patterns I found and that changes how we might want to think about the private commons and then therefore like the regul regulatory landscape or whatever like for example maybe everyone should be be able to download their data you know like stuff like that talk more about that. So first just a general thing is I tend to be call it more regulatory cautious than regulatory uh \>\> anti-regulation \>\> anti-regulation right um because it is generally true that when you get into oh well I should act as a regulator you then I I I got over a degree of certainty that I can issue a you shall do this and not this and actually enshrine the past against the future and there's all kinds of benefits to getting to the future and that's why regulatory caution So I tend to say look when you start having the impulse that maybe there should be regulation you should start with well how do we measure the questions that we're worried about as harms because if we can get those measurements then we can track it and see do we really have those harms are they really increasing and then maybe what kinds of innovations might might bring down those harms relative to our benefits and then allow that dynamic progression of innovation and invention and risk-taking all of which is very important and a lot regulators go well if I like you to take risk and something bad happens then it's on me and it's like well but like life is risk and by the way by taking some risk we can figure out things like cures for diseases and and other and we need we need to be figuring out the cures for diseases so um so anyway that's first general stance on regulation ne next is private commons which is you know part of what we actually seeing in these build of build up of these big tech companies counter or a bunch of, you know, surveillance capitalism and other kinds of critiques is is actually these \>\> I like that you included that in the book, by the way. \>\> Yeah, exactly. Because because like look, it's again, it's almost like framing like if you frame it as the agency is, oh, it's surveilling capitalism versus it's no no actually it's building up a private com benefit from enables me, right? Because you go, well, like Google Maps allows other people to figure out where your house is. And by the way, Google Maps allows my friends to figure out where my house is and come visit. Yes. Right. So um so that kind of framing I think is very important and then these data commons then are a kind of a micro instance of how super agency is created. Not only because you know I have a private common, you have a private common or private data sources, but then as it learns from these together, you get automatic tagging on your pictures and you can go, oh, this picture I took of Dan, I can share it with him, you know, etc., etc. And all this stuff adds in a lot of important human connectivity framework, uh, richness to our lives, ability to make decisions or find things in good ways. And um now that being said, we want to enable that private commons as much as possible. So you go well look in general most of the incentives for these corporations is very pro-social, pro like being good for society because it's like oh I'm building all these features that engage you, cause you to uh take pictures, cause you to store them, etc. And it's like great. But on the other hand, some of my incentives are and you do them only with me and you don't do them with anyone else. It's like well actually maybe occasionally in a very specific way we intervene and say no no no um part of my rights with my data is when you are uh storing a whole bunch of my data I have the right to move that as I like and you must facilitate that. \>\> Yeah. Totally. I'm I'm about to go off-roading because you're making me think of something. um which is uh like a more specific example that I think you might be interested in about like how this how this um might play out and I'm cur I'm curious your take on. So, um I don't know if we've talked about this before, but I have OCD. Um and there are treatments for OCD. Like you can you basically like manage it. Like it's it's fine. Um but like it's still not great. You don't cure it, right? And um one of the things I've been doing recently is I just have you used \>\> Windsurf. \>\> Windsurf is like a cursor competitor and it they just has like an agent thing. You can just be like build this build this thing and it'll build it. It's the coolest thing ever. Um, and so it just makes weekends a lot like really fun because I'm just like, "Oh, I have I have an idea." So I I had this thing that I've been doing where I just I built a little app in Windsurf. And um, every day I take a Ybox test, which is the OCD assessment. Um, and then I upload a screenshot of my Whoop data and then I also take a little video of myself and I just do like a daily log where I don't talk about like how I'm feeling. I just like talk about my day. Um, and I upload it and there's an a there are some APIs where it it embeds my face and my vo vocal tone. And what I'm trying to do is see if I can predict my OCD symptoms from my facial, vocal, and whoop data. Um, and uh, and like that's something that like Whoop, for example, they have an they have API, but like it doesn't really give you my like full all my data, and I need that in order to like make this work. Um and and this is the kind of thing where uh there's a whole there's like thousands of types of diseases or conditions or whatever where um if you enable people to do this, we would make so much more progress on like solving them. Like I don't know what I'm going to find when I when I figure this out. And and I think what I'm what I'm likely to find, which gets into my whole like soap box about science, uh which which we've talked about before, um is is that there's not one variable where it's like this makes your OCD worse. It's like actually there's like thousands of different things that like interact together and different combinations that like I should pay I should probably pay attention to, but they're all like weekly interacting. And yeah, I think that that kind of thing is like it's so important to the future and it it makes me feel like um there was this whole like quantified self movement in in Silicon Valley for a long time and it's like I feel like quantified self is like it's it's going to happen finally. It's like a thing now. Um what do you think? Well, so I do think that it's precisely this kind of data that if you have the ability to have kind of to shift it around as you need and want as an individual can create great things. So I think you know Whoop is one version of quantified self, glucose monitoring is another one. The you know things that you do with the various watches is more and I think that we are going to have all that and actually by the way having that um stream of data that you can then index to other data that you might have. For example, what did you eat for lunch? Um are you traveling? Where did you go? How did you uh like did you planes, trains, automobiles, etc. and and I think being able to query that together really really matters and and so I think it's a um a we are moving uh we are moving I think we're at this point decisively moving more and more towards quantified self but um you know part of it is is that enabling of of what I can do with it and therefore how it can really benefit me and by the Like for example, when you get to it, like one of the things that uh one of my doctors noted is that he can uh if he has access to the Whoop data and sees what's happening, he can look at someone's um you know, Whoop performance of the last I forget how many days, some number of days, he can tell whether or not you're about to get COVID. \>\> Right. Yeah. Yeah. Yeah. Yeah. \>\> Right. And so that's the kind of thing where it's actually that's really useful. \>\> Yeah. so important and there's probably, you know, there maybe there's one thing that he's looking at, but it's probably like he just sort of squints at it and gets like an intuitive sense of of of what it is, which is a thing that AI is like really quite good at because like predicting a um a sequence of let's say Whoop data or predicting uh or like classifying a sequence of Whoop data or let's say predicting a sequence of OCD data where it's like you know uh here's what I did today and here's my Ybox score or whatever. I think that that problem is incredibly similar to predicting the next token in a in like in language and we're we're just getting to the place where we're starting to like apply that in other places. Um, so you know, I think that there's this really interesting um shift that that happened in AI that I think you can sort of start to see it in first in philosophy where um we're going from uh like a a search for like universal truths and universal definitions in philosophy starting with like Kant really um you're starting to say like okay there's a limit to like our our ability to like grasp things intellectually and then there are all these like different flowerings of that. So like there's, you know, the transcendentalist and the American pragmatist, there's like the later Vickingstein, like there's there's the whole continental philosophy and like, you know, postmodernism or whatever. Uh, which regardless of what you think of that, it sort of is all in that all in that vein. Um and I think you can see this that same shift in the history of AI where um we started in a very like um um Socratic platonic aligned um mode in symbolic AI where we're trying to define um the underlying theory of intelligence and defining it as a set of symbols and their relations and each symbol has has some sort of semantic meaning. And what we found is like that doesn't really work. Like it's very brittle. Like it works in certain well- definfined circumstances, but it's really brittle and it suffers from um computational explosion. Like you have to have a frame of reference to start with before you can like start manipulating symbols otherwise it's it's it's um too expensive. And then we shifted to like subsyolic AI um and uh where we're we're just basically fuzzy matching patterns and and and and uh bringing to bear many many thousands of different rules um based on the context uh partially fitting them based on the context. And I think that that it's possible that that shift will also occur in other areas of the world where we're sort of doing the equivalent of like symbolic AI because there's nothing better. Um, and now we can use data and AI and that kind of stuff to like to to yeah to to make progress. \>\> Well, I think you know and we may talked about in the previous podcast that my undergraduate major was symbolic systems. \>\> Yes, I do. \>\> Yes. And um part of when I was doing that because the theory was we are symbol processors that we reason and think in symbols. We we consume it through you know languages and books etc. All of which is very interesting and bunch of different parallels between different symbolic systems. Um but I was also uh you know kind of within that very early movement which was called connectionism which was kind of this notion of well symbols are important but if you only had a symbolic theory you're probably going to radically underperform your modeling of what our intelligence is your ability to construct tools or intelligences etc. And so that sub symbolic AI and like for example what is what are the things that lead to us becoming like getting a mastery of concepts um or since we you know mentioned Vickinstein following a rule in language is I think uh really key. Now I think part of if I project if I project out the future for the next 10 years I think a major part of this is going to be this kind of like play between probabilistic models and symbols and with the current LLM transformers we have one but I think we're going to have more kind of as it were technologies um mathematical description skill sets about how probabilistic modes and symbols can come together. And I think that's among the things when you say what is currently not visible within the next generation of AI, I think there's going to be some stuff in there. \>\> Can you give me an example of what that might look like? Well, a simple one that's out of the past and I'm not saying this is the right one, but is like um using symbols but with Beijian probabilities estab you know kind of applied. So you say oh well I have the belief that we are uh that that that we're having philosophical insight you know in this podcast but it's probability 90%. Yeah, \>\> right versus probably 100%. \>\> Right. you know, like you know, that kind of thing. \>\> Only 90%. \>\> Well, I'm just giving some range for humility and all the rest. Um, and so, um, but you might also say like, I have the I have the belief that the weather tomorrow will be good \>\> for taking a hike and I believe it at 80%. Yeah. \>\> Is the 01 paradigm a example of this like um training it on um like math pro like any any kind of problem where uh it's it's based based on discrete steps that can be verified. Um I don't think it's unusually that from a lot of the transformers but I do think that the question is is you know you do these chains of thought you have a fitness function on the chain of thought and you're making a prediction not just the next token but is this chain of thought out of multiple chains of thought which of these ones are the right ones and that fitness function itself might have that probabilistic characteristic. Yeah, cuz cuz I don't want to um I like to cast dispersions on symbols. \>\> With symbols, I might add. \>\> With symbols. I love symbols \>\> to cast those dispersions. But that's mostly cuz I it's like um it's sort of obviously like a little bit rhetorical like uh because I think we're so symbol heavy um in how we think about things. Um, and I guess my I'll at least my model right now and that's open to revision is like symbols are important but they are important when they arise out of a sub subsy symbolic architecture. Um, and so, uh, all that means is that when you're when you're doing something like, for example, um, trying to pick apart your thought process, um, there's a limit to that like picking apart where you're getting down to the like the symbolic level for yourself, but underneath that are like there's, you know, thousands and thousands of subsymbolic things that you're not going to be aware of. and know and realizing how the symbolic level arises out of the subsy symbolic level is like a really important um uh way to approach the world especially if you're someone who has a tendency to philosophize because you're like it's the sort of vicinian like therapeutic approach to philosophy which is like you need to find a a philosophical stance that allows you to give up philosophizing. \>\> Yes. Although I think even maybe he realized that it wasn't that it was never ended and you gave up cuz you know obviously early Vickenstein I solved it. I'm done. I'm going to go teach. Oh wait a minute. \>\> Talk about ego. \>\> I got to go back and redo this. \>\> Talk about agency. \>\> Yes. Exactly. Which is the reason why he, you know, died as a philosophy professor. \>\> Yeah. Um every time I thought I was out they pull me back in is Vickenstein. \>\> Vickenstein's godfather. \>\> Yeah. Um let's see what else. what else I have to talk about with you. Um, we've gone a little bit into um, we've gone a little bit into science versus engineering, which I do think like, um, when we talk about dynamic tensions, um, you could probably think of everything post Newton as like the era of science. Um, and I think there's a way of looking at the era of, um, of AI that is starting to work as like turning science problems into engineering problems. and that it may usher in more of an era of engineering um or like lean us more in in that um in that direction which has a bunch of other like um Jackson Pollock like you you can like twist the knobs on other dynamic tensions in this sort of engineering direction which is like we're a little bit more pragmatic um we're a little bit less concerned with um uh uh maybe with like fully explicit causal explanations for things and maybe uh more okay with um uh lots and lots of little correlations um stuff like that. \>\> By the way, I obviously love the Jackson Pollock um and I would also add since you've done that one already, another metaphor which is thinking about or framework thinking about like we're actually moving less from thinking about these thoughts as deduction and more as induction and abduction. \>\> Yeah. What is abduct abduction? I never remember. \>\> Oh, it's the best theory. \>\> Okay. \>\> It's kind of theory that models the evidence. Interesting. Okay. Interesting. Um, so but it requires like uh some sort of creative process to come up come up with the theory rather than like coming up with it from the data itself. \>\> Yeah. A little bit like if you if you contrast induction and abduction. Induction is I I look at all the data points and I model the curve and abduction is based on some um other information about the world or priors. I go, here's a model which might actually be different than what you come up with an induction. It's one of the reasons why I actually think both like frankly all three, but part of when we're good thinking beings, we apply all of them. \>\> Yeah, that's that is that is the thing that I love about um AI is like you don't have to decide which way to approach problems. It just like figures out it has many different approaches and it just figures out the one that's that's best um or uses many different ones at the same time, which I I think is sort of like the failure point of like a lot of philosophizing. So, like coming up with um uh a like moral theory, right? Like we've never found one that like has no holes in it. Like um and and when we do think that we found it, um you you end up with all these like weird things. So like um uh uh effective altruism is a great idea, but then you like take it too far and then you like become SPF. Like it's \>\> Yes. Um, and so but but then it it feels really wishy-washy to be like, well, I applied the moral rule that like made the most sense in this circumstance, but that's how language models work um for for lang languages or solving problems. And and I think um having some sort of um uh diverse partially explicit but like also mostly implicit set of like moral theories or or whatever you want to call it um is probably for the best. It's probably the best way to to to be in the world which is like fundamentally sort of philosophically unsatisfying from the perspective of trying to make everything explicit but is also kind of beautiful and and um uh goes back to the stance towards uncertainty the like sort of artistic entrepreneur entrepreneurial stance toward uncertainty. \>\> Yeah. Look, I think it's critical that we're deriving theories of the world, theories of morality, theories of ourselves, theories of what the what we should be doing with our lives and work. But it's also critical to think of that those theories as dynamic and being updated. And part of the updating is not just, you know, kind of, oh, look, I got some new data. That's important, too, but also the way we think about it, the way we learn from each other. And all of that leads to kind of a you know best judgment kind of circumstance. And by the way, that's how science progresses too. Yeah. So people are uh you know they've they've listened to this discussion. Um they're they're familiar with with some of the ideas of the book. Um and they're probably thinking about like should I go should I go get it? Um, and I'm curious like um, yeah, are there other are there any other like takeaways that you think are really important for people to know about about what your stance is on these topics? \>\> Well, so in a sense I wrote the book for two audiences. Um, one audience is anyone who has any AI curiosity or skepticism because the thought is here's a set of lenses to think about the reason why this is very humanism and humanity positive and why it is important to have a a kind of a theory of agency that is this is what we can accomplish. This is what would be really good and how to work towards it. And so whether you're a skeptic or whether you're curious, all of that I think plays into it. Now I also wrote it for technologists because I wanted technologists who were inventing this to be thinking about human agency as almost like a design principle to think about you know individual agency collective super agency as ways of ah if that's one of the fitness functions and goods of what we're trying to do that uh may make certain design decisions deployment decisions much more effective and much more humanist. How does that work? Because I think a lot of the framing that we've approached agency with so far is a is a much more like we've talked about it as like an internal aesthetic sense rather than something that you can build into a product. So, how would that how would that change um a technologist's fitness function for what they're building? Well, that was a little bit of the reason why we highlighted the chat GBT iterative deployment as an example because you know GBD 3.5 had existed for like a year before chat GBT and yet you launch GVD and all of a sudden people could access it. They could do stuff with it. They could make it happen. And so it's making that kind of uh affordances available to people so they can engage with it and that they're interested in engaging with it and that they're easy to do an iterative pattern of engaging is one of the things that comes out of that. I also think that you know if I were to uh you know be able to say hey everyone make sure you're doing this as well as this is obviously when we build a lot of technology we tend to think about the easiest simple path is a form of human replacement. So you go okay we don't need customer service agents we'll just do this. What I think it's also super important is to think about like that co-pilot that human amplification and making it more as per the book impromptu amplification intelligence. I think that that um that's also something to be thinking about with a kind of a super agency design lens. \>\> Yeah. One of the things that that I think about that as as a lens for for trying to figure out places to build AI that increases agency is that it um reduces the cost of certain um intelligence services that people pay for right now that use humans. Um so that not so that they don't use humans that other people who can afford them um can get the same thing. So like a really simple example is I run a media company. Uh, I don't write all of the like YouTube headlines and all of the like um uh descriptions and and all that kind of stuff. Like I have a ghost writer that I work with who's super talented. Um, but if you're just starting out like you can't afford that. Um, and but you can use CHBT or Claude or we have an internal incubation we did called Spiral that does this and so it lowers the cost for people to like have a lot of the leverage that I get because I have more money in the whole organization. There are lots and lots of areas like that that go beyond creative stuff that I think are are really important and will be really useful for people. \>\> Yeah. I think part of the thing I think with AI and its amplification intelligence is I think it raises the bar like raises the capability of the folks who have less like I don't have access to lead school I can still learn. I don't have access to a ghostriter I can still write some marketing copy. But by the way, it also raises the well, you have a ghostriter who's really good. That ghostriter learns to use chat GBT and then also can do that in a much stronger and better way. And I think that's good. Like I think amplifying across the board is actually a good thing. \>\> Yeah, totally. Um, let me think about if there's anything else for us to discuss. \>\> Well, I'm sure there will be. \>\> I'm sure there will be. Um, this is great. Um, thank you so much for doing this. Uh, I had a great time. Uh, I can't wait for the book to come out and I can't wait for another one. I would love to have you on again. \>\> Delight. I look forward to our next conversation. \>\> Oh my gosh, folks. You absolutely positively have to smash that like button and subscribe to AI and I. Why? Because this show is the epitome of awesomeness. It's like finding a treasure chest in your backyard, but instead of gold, it's filled with pure, unadulterated knowledge bombs about chat GPT. Every episode is a roller coaster of emotions, insights, and laughter that will leave you on the edge of your seat, craving for more. It's not just a show, it's a journey into the future with Dan Shipper as the captain of the spaceship. So, do yourself a favor, hit like, smash subscribe, and strap in for the ride of your life. And now without any further ado, let me just say Dan, I'm absolutely hopelessly in love with you.