Advertise on podcast: Increments

Vaden has been annoyed about the Hugging Face incident for weeks, and the only known cure is yelling about it on the podcast. Ben, who as usual is more even-headed and objective about the incident (incidentally, he also writes the episode descriptions) plays devil's advocate. Was there ever really a sandbox? Did the agents cheat, or did they just do what they were told? Is terrifying the public good for an IPO? Multiple analogies enter the fray, zero minds are changed, and Saturday Night Live gets the last word. We discuss What actually happened in the Hugging Face incident (and what is ExploitGym?) What Project Glasswing and Mythos already told us about AI and cybersecurity Is a sandbox with Artifactory access still a sandbox? Did the agents cheat, or just follow their prompt? Negligence, PR stunt, or both? Going through OpenAI's timeline How independent is an "independent" investigation? Does anthropomorphic language help you predict what AI will do? Grenades, knives, and yogurt: parenting styles for package managers References The Hugging Face incident and the road ahead (OpenAI blog post): https://openai.com/index/hugging-face-incident-and-the-road-ahead/ OpenAI's technical report on the incident (PDF): https://cdn.openai.com/pdf/67869394-cb91-4c12-888c-5cbd85c7814c/OpenAI-Hugging-Face%20Incident-Technical-Report.pdf Making Sense #494, "A Coin Toss for the Future" (Sam Harris with Ryan Greenblatt): https://www.samharris.org/podcasts/making-sense-episodes/494-a-coin-toss-for-the-future Redwood Research on the incident: https://blog.redwoodresearch.org/p/the-openaihuggingface-incident-redwood Dwarkesh Podcast with Ajeya Cotra, "Inside the OpenAI agent swarm that hacked Hugging Face": https://www.youtube.com/watch?v=X50zezLFWWI&pp=ygUVZHdhcmtlc2ggaHVnZ2luZyBmYWNl Melanie Mitchell, "Misleading Metaphors and Real Risks": https://aiguide.substack.com/p/misleading-metaphors-and-real-risks Anthropic's Project Glasswing: https://www.anthropic.com/glasswing JFrog Artifactory: https://jfrog.com/artifactory/ Our episode with Melanie Mitchell: https://www.incrementspodcast.com/45 Socials Follow us on Twitter at @IncrementsPod, @BennyChugg, @VadenMasrani Come join our discord server! DM us on twitter or send us an email to get a supersecret link Become a patreon subscriber here. Or give us one-time cash donations to help cover our lack of cash donations here. Click dem like buttons on youtube We urge you to urge us to stop. Send your urges over to [email protected]

At 3am, tossing and turning in your bed, have you ever wondered why this generation is full of a bunch of fragile sissies? Journalist Abigail Shrier has, and her answer is: Therapy. Too much God d*** therapy. Is she right? Ben and Vaden read her 2024 book on the subject, Bad Therapy, and discuss it, including the good, the bad, and the preposterous footnotes. Buy the book here https://www.amazon.com/dp/0593542924 and check out the rest of Shrier's work here https://www.abigailshrier.com/. We discuss Iatrogenesis and whether there are negative side effects of therapy Why overly focusing on your feelings can lead to unhappiness Should kids be asked about suicidal ideation? Some problems with Shrier's analysis For God's sale, are kids fragile, or are they not? References Bad Therapy: https://www.amazon.com/dp/0593542924 Irreversible Damage: https://en.wikipedia.org/wiki/Irreversible_Damage Dr. K's youtube channel: https://www.youtube.com/c/HealthyGamerGG Part 1 of Decoding the Gurus series on Dr K: https://decoding-the-gurus.captivate.fm/episode/dr-k-part-1-oped-ayurvedic-medicine-vs-nerfed-western-allopathy Let Grow: https://letgrow.org/ Socials Follow us on Twitter at @IncrementsPod, @BennyChugg, @VadenMasrani Come join our discord server! DM us on twitter or send us an email to get a supersecret link Become a patreon subscriber here. Or give us one-time cash donations to help cover our lack of cash donations here. Click dem like buttons on youtube Would you rather go to therapy or fight in WWII? Tell us at [email protected]

After Jason Arday, the second biggest topic of conversation has of course been the rise of the neuro-grifting guru. And for good reason! Neuroscience has been causing confusion lately. What is the hippocampus up to in there? Is your pre-post-medial-cerebrum-corta-bellum even working properly? What does neuroscience say about whether you should start each day with a slap in the face? Being incapable of answering these questions ourselves, we're bringing on bonafide neuroscientist and substacker Tommy Blanchard to help us navigate this terrain. Check out Tommy's website and substack, Cognitive Wonderland. We discuss Models of decision-making in neuroscience The common currency model What do we actually know about the brain? What constitutes an explanation in neuroscience? What neuroscience is put on a pedestal Neuro-grifting Gurus, aka Huberman et al. References Tommy's post about his research appearing in the NY Times: https://cognitivewonderland.substack.com/p/scientists-hope-gambling-monkeys Socials Follow us on Twitter at @IncrementsPod, @BennyChugg, @VadenMasrani Come join our discord server! DM us on twitter or send us an email to get a supersecret link Become a patreon subscriber here. Or give us one-time cash donations to help cover our lack of cash donations here. Click dem like buttons on youtube Which area of your brain lights up when you consider sending an email to [email protected]? Special Guest: Tommy Blanchard.

We've spent a long time bashing Bayesian epistemology on the podcast. Unfortunately, this is sometimes taken to mean that we bash all things Bayesian. But our God-fearing listeners will be pleased to learn that we do not bash all things that the Reverend Thomas Bayes stood for. Bayesian statistics, when appropriately separated from epistemology, has many things to teach us. And instead of taking our word for it, we're bringing on an expert in this field: Alex Andorra, a research scientist at Meta and host of the Learning Bayesian Statistics podcast. The Gods Smile. Check out Alex's website and the Learning Bayesian statistics podcast. We discuss 538 and election forecasting Sports analytics and why Messi is objectively the best What is Bayesian statistics? What is the difference between Bayesian and frequentist statistics? What is the allure of Bayesian statistics? Are priors best thought of as beliefs? How do critical rationalists make decisions? References Learning Bayesian Statistics: https://learnbayesstats.com/ Silver Bulletin: https://www.natesilver.net/ Vaden's blog post on decision making: https://vmasrani.github.io/blog/2020/vaden_second_response/#what-other-options-to-decision-theory-are-there # Socials Follow us on Twitter at @alex_andorra, @IncrementsPod, @BennyChugg, @VadenMasrani Come join our discord server! DM us on twitter or send us an email to get a supersecret link Become a patreon subscriber here. Or give us one-time cash donations to help cover our lack of cash donations here. Click dem like buttons on youtube Send us an email because you have - no, because you want to - over at [email protected]. Special Guest: Alex Andorra.

It's only taken us 104 episodes of talking incessantly about Karl Popper to get one of the most important and popular criticisms of falsification: The Duhem-Quine thesis. Should we wrap the podcast here? Is it game over? We discuss Does AI need more economic thinking? The role of background knowledge in conversation The allure of verificationism The Duhem-Quine thesis The role of competing theories in combatting the thesis Are conspiracies the price we pay for independent thinking? References Michael I. Jordan on ML Street Talk: https://www.youtube.com/watch?v=AREWYbVtX64 Our conversation with Ben Recht: https://www.incrementspodcast.com/91 Quotes People involved in a fruitful critical discussion of a problem often rely, if only unconsciously, upon two things: the acceptance by all parties of the common aim of getting at the truth, or at least nearer to the truth, and a considerable amount of common background knowledge. This does not mean that either of these two things is an indispensable basis of every discussion, or that these two things are themselves ‘a priori’, and cannot be critically discussed in their turn. It only means that criticism never starts from nothing, even though every one of its starting points may be challenged, one at a time, in the course of the critical debate. - C&R, Chap 10 While discussing a problem we always accept (if only temporarily) all kinds of things as unproblematic: they constitute for the time being, and for the discussion of this particular problem, what I call our background knowledge. Few parts of this background knowledge will appear to us in all contexts as absolutely unproblematic, and any particular part of it may be challenged at any time, especially if we suspect that its uncritical acceptance may be responsible for some of our difficulties. But almost all of the vast amount of background knowledge which we constantly use in any informal discussion will, for practical reasons, necessarily remain unquestioned; and the misguided attempt to question it all—that is to say, to start from scratch—can easily lead to the breakdown of a critical debate. (Were we to start the race where Adam started, I know of no reason why we should get any further than Adam did.) - C&R, Chap 10 Socials Follow us on Twitter at @IncrementsPod, @BennyChugg, @VadenMasrani Come join our discord server! DM us on twitter or send us an email to get a supersecret link Become a patreon subscriber here. Or give us one-time cash donations to help cover our lack of cash donations here. Click dem like buttons on youtube What kind of thinking does this podcast need? Tell us at [email protected].

Did you ever see the video of two regular people trying to fight Eddie Hall, the strongest man on earth? Ben and Vaden bring that same energy to this conversation, where they gang up on Robert Wright on the subject of his forthcoming book The God Test: Artificial Intelligence and Our Coming Cosmic Reckoning. Bob is a sport about it (well, there was no choice really: we have an advanced copy of his book that we're holding hostage) and there are only one or two black eyes by the end. Subscribe to Robert Wright's podcasting empire NonZero on youtube and substack. And go buy one of his many books. We discuss Whether Bob was the OG podcaster Is there a directionality to evolution? Where Bob is on the AI fear and awe spectrum Whether LLMs have decoded the meaning in language Is AI best viewed as a tool or as superintelligence? AI-as-slingshot analogy References The God Test by Robert Wright Nonzero by Robert Wright William Paley and his pocket watch The Blind Watchmaker by Richard Dawkins Our conversation with Scott Aaronson Socials Follow us on Twitter at @RobertWrighter, @IncrementsPod, @BennyChugg, @VadenMasrani Come join our discord server! DM us on twitter or send us an email to get a supersecret link Become a patreon subscriber here. Or give us one-time cash donations to help cover our lack of cash donations here. Click dem like buttons on youtube What's the directionality of this podcast? Tell us at [email protected]. Special Guest: Robert Wright.

Turning on our own! Vaden sits down with Bruce Nielson and Peter Johansen to discuss some of the bad memes that run rampant in the online "crit-rat" community, in an attempt of some intra-tribal error-correction. Ben, scared of the retribution, runs away with his tail uncomfortably close to his thighs. Follow Bruce and Peter's work on the Theory of Anything Podcast. Sign up for the Theory of Anything podcast Patreon here, and Check out Peter's Many World of David Deutsch facebook group here. We discuss Bruce and Peter's origin stories Anarcho-capitalism and its (pseudo) connections to Popperian philosophy CritRat disbelief in mental illness What does the Universal Explainer hypothesis imply about mental illness and IQ? Is Anarcho-capitalism a foundationalist and utopian philosophy? Why are so many crit-rats Trump supporters? See Also Theory of Anything Podcast #136: Michael Golding on Mental Illness Theory of Anything Podcast #82: Popper's Ratchet Socials Follow us on Twitter at @IncrementsPod, @BennyChugg, @VadenMasrani, @bnielson01 Come join our discord server! DM us on twitter or send us an email to get a supersecret link Become a patreon subscriber here. Or give us one-time cash donations to help cover our lack of cash donations here. Click dem like buttons on youtube I before E except after B? Send your best guess as to the spelling of Bruce's last name (no one knows) over to [email protected]. Special Guest: Bruce Nielson.

Wasn't Popper a falsificationist? Then why did he try to develop ideas about corroboration and versimilitude - the extent to which a theory was closer to truth than another theory? Isn't this verging dangerously close to verificationist territory? In our fourth ep on Chapter 10 in C&R, we wrestle with Popper's treatment of verisimilutude, both the formal and informal versions. Did the project fail? Was Popper out of his mind? Does this invalidate everything? We discuss Murders with ball-peen hammers Walking the line between verification and falsification Is science only after truth? Verisimilutude and its formalization Why the formalization fails Popper's three requirements for the growth of knowledge Popper's ratchet and the no ad-hoc rule Quotes Like many other philosophers I am at times inclined to classify philosophers as belonging to two main groups—those with whom I disagree, and those who agree with me. - C&R, page 309 I shall give here a somewhat unsystematic list of six types of cases in which we should be inclined to say of a theory t1 that it is superseded by t2 in the sense that t2 seems—as far as we know—to correspond better to the facts than t1 , in some sense or other. t2 makes more precise assertions than t1 , and these more precise assertions stand up to more precise tests. t2 takes account of, and explains, more facts than t1 (which will include for example the above case that, other things being equal, t2 ’s assertions are more precise). t2 describes, or explains, the facts in more detail than t1 . t2 has passed tests which t 1 has failed to pass. t2 has suggested new experimental tests, not considered before t 2 was designed (and not suggested by t1 , and perhaps not even applicable to t1 ); and t 2 has passed these tests. t2 has unified or connected various hitherto unrelated problems. - C&R, page 315 Let me first say that I do not suggest that the explicit introduction of the idea of verisimilitude will lead to any changes in the theory of method. On the contrary, I think that my theory of testability or corroboration by empirical tests is the proper methodological counterpart to this new metalogical idea. The only improvement is one of clarification. - C&R, page 318 Socials Follow us on Twitter at @IncrementsPod, @BennyChugg, @VadenMasrani Come join our discord server! DM us on twitter or send us an email to get a supersecret link Become a patreon subscriber here. Or give us one-time cash donations to help cover our lack of cash donations here. Click dem like buttons on youtube How many chromosomes does diethyl-methyl pentophosphate have, exactly? Tell as at [email protected]

100 episodes! To celebrate, Vaden tries to get personal with Ben, while Ben dodges his questions and wants to know how Vaden feels about incest. All in all, a pretty typical episode. The questions From Vaden to Ben: How is your side hustle going? Who are some of your major influences outside of Popper? How has the Popperian worldview influenced your day-to-day? What is the life of a nomadic academic like? What would you say to people who are considering mathematics as a career? Which charities do you recommend? From Ben to Vaden: How do you feel about looksmaxxing? Thoughts on medical assistance in dying? Ethics of Alex Honnold free soloing Taipei 101? Thoughts on Nation-Buiding? Incest - into it? Episode References #22 - Thinking Through Thought Experiments #66 - Sex Research, Addiction, and Financial Domination (w/ Aella) #58 - Ask Us Anything V: How to Read and What to Read #70 - ... and Bayes Bites Back (w/ Richard Meadows) #76 (Bonus) - Is P(doom) meaningful? Debating epistemology (w/ Liron Shapira) References Angus Deaton debates Abhijit Banerjee: https://nyudri.wordpress.com/initiatives/deaton-v-banerjee/ Christopher Hitchens and Robert Wright Sam Harris and Garry Kasparov Private Empire: ExxonMobil and American Power The Bomb in My Garden: The Secrets of Saddam's Nuclear Mastermind Socials Follow us on Twitter at @IncrementsPod, @BennyChugg, @VadenMasrani Come join our discord server! DM us on twitter or send us an email to get a supersecret link Become a patreon subscriber here. Or give us one-time cash donations to help cover our lack of cash donations here. Click dem like buttons on youtube What is your favorite form of ince... actually nevermind, too much. Just email us at [email protected].

Ben is hesitant to talk too much about politics on the podcast. So, naturally, we're having a political debate. We have on a pseudonymous guest to talk Trump, MAGA, and what's wrong with the left and the right. We discuss Trump's politics as trial and error The dynamics of MAGA ICE activity in Minnesota Illegal immigration Would Popper have supported Trump? What does "incrementalism" mean in politics? Socials Follow us on Twitter at @IncrementsPod, @BennyChugg, @VadenMasrani Come join our discord server! DM us on twitter or send us an email to get a supersecret link Become a patreon subscriber here. Or give us one-time cash donations to help cover our lack of cash donations here. Click dem like buttons on youtube Cross the border into the increments discord and hit us up at [email protected].

"What is Truth?", said jesting podcasters, who then stuck around for an answer. Back at it again with The Conjectures and Refutations Series (part three) on Chapter 10: Truth, Rationality, and the Growth of Scientific Knowledge. Can we say what truth is, even if we can never be certain we've found it? If not, can we say that science is approaching truth? How would we ever know? And why are so many theories of truth untrue? We discuss Ben's early reflections on Abigail Shrier's book Bad Therapy Why did Popper feel the need to answer this particular "what is" question? Can asking "what is truth" be a demogogic and bad-faith question? The correspondence theory of truth vs The pragmatic theory of truth vs The coherence theory of truth Alfred Tarski's formalization of the correspondence theory of truth Are there problems with the correspondence theory? The disagreement between Vaden and Deutsch on truth References Daniel Bonevac on the Correspondence theory of truth: https://www.youtube.com/watch?v=qlG_VaN1LHQ Tarki's 1944 paper on the semantic conception of truth Tarki's 1933 paper "On the concept of truth in formalized languages" Deutsch's 2022 talk on truth: Musings about Truth # Socials Follow us on Twitter at @IncrementsPod, @BennyChugg, @VadenMasrani Come join our discord server! DM us on twitter or send us an email to get a supersecret link Become a patreon subscriber here. Or give us one-time cash donations to help cover our lack of cash donations here. Click dem like buttons on youtube It would be both useful, coherent, and correspond to our happiness if you signed up for our patreon or discord. Hit us up at [email protected]

Two years without discussing effective altruism -- did you miss it? Not as much as Vaden, surely. And probably a right bit more than Ben. Well, we're back in the game with a spicy one. Was EA a front for AI safety from the beginning? Did the leaders care not a wit for global poverty? Is Ben going to throw himself out window if Vaden keeps this up? We discuss Feedback on our introspection episode The motives of the EA founders The felicia forum Is this a conspiracy theory? EA's strategic ambiguity Bostromism, transhumanism, and AI safety EA funding The public/core divide and the funnel model Quotes new effective altruists tend to start off concerned about global poverty or animal suffering and then hear, take seriously, and often are convinced by the arguments for existential risk mitigation - Will MacAskill Existential risk isn’t the most useful public face for effective altruism – everyone inc[l]uding Eliezer Yudkowsky agrees about that - Scott Alexander, 2015 Utilitymonster: GWWC is explicitly poverty-focused but high impact careers (HIC) is not. In fact, hardcore members of GWWC are heavily interested in x-risk, and I estimate that 10-15% of its general membership is as well. I’d take them seriously as a group for promoting utilitarianism in general. I’m a GWWC leader. [Redacted]: but HIC always seems to talk about things in terms of “lives saved”, ive never heard them mentioning other things to donate to. […] Utilitymonster: That’s exactly the right thing for HIC to do. Talk about lives saved with their public face, let hardcore members hear about x-risk, and then, in the future, if some excellent x-risk opportunity arises, direct resources to x-risk. - From felicia forum. References Gleiberman's paper: https://medialibrary.uantwerpen.be/files/8518/61565cb6-e056-4e35-bd2e-d14d58e35231.pdf Old EA wikipedia page (web archive): https://web.archive.org/web/20170409171350/https://en.wikipedia.org/wiki/Effective_altruism Old CEA webpage (web archive): https://web.archive.org/web/20161219031827/https://www.centreforeffectivealtruism.org/fundraising/ Socials Follow us on Twitter at @IncrementsPod, @BennyChugg, @VadenMasrani Come join our discord server! DM us on twitter or send us an email to get a supersecret link Become a patreon subscriber here. Or give us one-time cash donations to help cover our lack of cash donations here. Click dem like buttons on youtube Let us funnel you into the core group of super secret patreon supporters. Send us an email at [email protected]

Happy Christmas and Merry Festivus y'all! Today we're releasing a patreon episode, as both of us are away on vacation with the family for the holidays. In this episode we have a meandering discussion about parenting, Robert Kegan's four stages of development, the limits of introspection, and relationship counseling. We discuss Advice for new fathers Vaden comes out to the world (about snowboarding) Robert Kegan's four stages of development The limits of introspection Countertransference The show Couples Therapy with Dr. Orna Guralnik # Socials Follow us on Twitter at @IncrementsPod, @BennyChugg, @VadenMasrani Come join our discord server! DM us on twitter or send us an email to get a supersecret link Become a patreon subscriber here. Or give us one-time cash donations to help cover our lack of cash donations here. Click dem like buttons on youtube Introspect a little, and tell us why you haven't signed up to our Patreon feed over at [email protected]

After a long hiatus where we both saw grief counsellors over our fight about Popper's theory of content in the last C&R episode, we are back. And we're ready to play nice ... for about 30 seconds until Vaden admits that two sentences from Popper changed his mind about something Ben had arguing for literally years. But eventually putting those disagreements aside, we return to the subject at hand: The Conjectures and Refutations Series: Chapter 10: Truth, Rationality, and the Growth of Scientific Knowledge (Part II). Here all goes smoothly. Just kidding, we start fighting about content again almost immediately. Where are the guests to break us up when you need them. We discuss Why Vaden changed his mind about "all thought is problem solving" Something that rhymes with wero horship Is Popper sloppy when it comes to writing about probability and content Is all modern data science based on the wrong idea? (Hint: No) Popper's problem-focused view of scientific progress How much formalization is too much? The difference between high verisimilitude and high probability Why do we value simplicity in science? Historical examples of science progressing via theories with increasing content Quotes Consciousness, world 2, was presumably an evaluating and discerning consciousness, a problem-solving consciousness, right from the start. I have said of the animate part of the physical world 1 that all organisms are problem solvers. My basic assumption regarding world 2 is that this problem-solving activity of the animate part of world 1 resulted in the emergence of world 2, of the world of consciousness. But I do not mean by this that consciousness solves problems all the time, as I asserted of the organisms. On the contrary. The organisms are preoccupied with problem-solving day in, day out, but consciousness is not only concerned with the solving of problems, although that is its most important biological function. My hypothesis is that the original task of consciousness was to anticipate success and failure in problem-solving and to signal to the organism in the form of pleasure and pain whether it was on the right or wrong path to the solution of the problem. In Search of a Better World, p.17 (emphasis added) The criterion of potential satisfactoriness is thus testability, or improbability: only a highly testable or improbable theory is worth testing, and is actually (and not merely potentially) satisfactory if it withstands severe tests—especially those tests to which we could point as crucial for the theory before they were ever undertaken. - C&R, Chapter 10 Consequently there is little merit in formalizing and elaborating a deductive system (intended for use as an empirical science) beyond the requirements of the task of criticizing and testing it, and of comparing it critically with competitors. - C&R, Chapter 10 Admittedly, our expectations, and thus our theories, may precede, historically, even our problems. Yet science starts only with problems. Problems crop up especially when we are disappointed in our expectations, or when our theories involve us in difficulties, in contradictions; and these may arise either within a theory, or between two different theories, or as the result of a clash between our theories and our observations. - C&R, Chapter 10 Socials Follow us on Twitter at @IncrementsPod, @BennyChugg, @VadenMasrani Come join our discord server! DM us on twitter or send us an email to get a supersecret link Become a patreon subscriber here. Or give us one-time cash donations to help cover our lack of cash donations here. Click dem like buttons on youtube Is "Ben and Vaden will fight about content" high or low probability? Tell us at [email protected]

The time has come for Vaden to defend his faith in the face of cold, hard scientific rationality. Will AI take over the world, automating away everything that makes humans distinct? Or can Vaden defend the church of just-ism, the radical belief that AI is simply "just a tool." Scott Aaronson, professor of computer science at UT Austin, goes to head to head against the zealotry. Check out Scott's website and his blog, Shtetl Optimized. We discuss Scott view's on education. Should we radically reform K-12? Is ChatGPT changing Scott's approach to teaching The religion of "justa-ism" Is AI just a tool? Is there any principle which lets us say that AI won't be as general as humans? Aaronson's thesis of Artificial Intelligence Computational universality vs explanatory universality The many-worlds interpretation of quantum mechanics Socials Follow us on Twitter at @IncrementsPod, @BennyChugg, @VadenMasrani Come join our discord server! DM us on twitter or send us an email to get a supersecret link Become a patreon subscriber here. Or give us one-time cash donations to help cover our lack of cash donations here. Click dem like buttons on youtube Have you been converted? Tell us at [email protected] Special Guest: Scott Aaronson.