Logo
    Search

    Tyler Cowen - Hayek, Keynes, & Smith on AI, Animal Spirits, Anarchy, & Growth

    enJanuary 31, 2024

    Podcast Summary

    • Investment and innovation involve a mix of rational and irrational factorsKeynes argued that investment results often disappoint expectations and people invest based on profit desire and satisfaction, while innovators may not fully capture social value of their creations.

      Learning from the discussion about the book "GOAT: Who is the Greatest Economist of All Time and Why Does It Matter?" is the idea that investment and innovation often involve a degree of irrationality or overoptimism. Keynes, who is quoted in the book, argued that the actual results of investments have often disappointed expectations, and that people are driven to invest not just by rational calculation but also by a desire for profit and satisfaction. This idea is still relevant today, as active investing and entrepreneurship continue to be driven by a mix of rational and irrational factors. Another point raised in the discussion is that innovators and entrepreneurs may not fully capture the social value of their innovations or creations. For example, Picasso's income may not accurately reflect the social value of cubism, and the same could be said for entrepreneurs and their businesses. Overall, the discussion highlights the complex and often irrational nature of investment and innovation, and the challenges of accurately measuring their social value.

    • Keynes on Long-Term Investment vs. Short-Term SpeculationKeynes believed long-term investment was risky and difficult, while short-term speculation could be profitable. His views may not apply to modern markets due to changed economic conditions.

      Learning from the discussion about Keynes' views on investment and markets is that for him, long-term investment based on genuine expectations was considered difficult and risky, and he believed that short-term speculation could be more profitable. However, his views were likely influenced by the specific economic conditions of his time and may not be applicable to modern markets. Keynes also expressed concerns about the increasing role of passive investors in the market and the potential for collusion among them, which could impact competition and the efficient functioning of markets. Overall, while some of Keynes' ideas may seem outdated or overconfident, they still provide valuable insights into the complexities of investing and the role of markets in the economy.

    • Striking the Balance: Optimal Trading and the Financial Sector's SizeIndividuals' trading levels depend on temperament, and the financial sector's size is a subject of debate. People's risk behavior varies, and intellectuals' contributions should be evaluated holistically.

      The optimal level of trading in financial markets might be a matter of individual temperament, and it can be challenging to strike the perfect balance between too much and too little. From a social perspective, it might be desirable for everyone to have the temperament to do what's socially optimal, but human emotions and inclinations make that an unrealistic goal. The financial sector's size relative to GDP or wealth is another topic of debate. While some argue that the financial sector should be smaller, others believe that its growth is not necessarily sinister. Milton Friedman's perspective on risk aversion suggests that humans are not inherently risk-averse or risk-loving but context-dependent. People's risk behavior can be seen as a form of mood management, and what seems risky to one person might not be to another. Ultimately, evaluating the contributions of intellectuals like Hayek requires considering their entire body of work, recognizing that not all risks pay off, and understanding that their less successful endeavors might still hold value in broader contexts.

    • Understanding Intellectuals' Contributions: The Importance of Their Extensive Bodies of WorkIntellectuals' value comes from their extensive bodies of work, which reveal their depth and breadth of knowledge, shape their perspectives, and contribute significantly to various fields. Their ability to navigate societal norms and effectively communicate ideas influences their impact on society and thought.

      While the major failures and successes of intellectuals may be the most noteworthy, the full scope of their work is essential for understanding their contributions to society and thought. The failures and successes are interconnected, shaping the intellectuals' perspectives and evolving ideas. For instance, Hayek's inability to establish a normative standard in the Constitution of Liberty showcases his lack of rigor but doesn't detract from his profound work. Similarly, Keynes's inconsistencies and adaptability reveal his strengths and weaknesses. The importance of an intellectual's extensive body of work lies in their depth and breadth of knowledge. Figures like John Stuart Mill, who wrote extensively on various topics, showcased their versatility and contributed significantly to various fields. The influence of intellectuals, such as Keynes, stems from their ability to navigate societal norms and effectively communicate their ideas. While some, like Turing, may have faced harsher consequences for their actions, others, like Keynes, could leverage their political power and astute career management to maintain influence. In essence, the long-term value of intellectuals lies in their ability to shape discourse, challenge ideas, and provide new perspectives, even if their specific conclusions are later proven incorrect.

    • The Role of Economic Growth in Society's Progress and Potential Future ChallengesEconomic growth has led to significant improvements, but potential future challenges like economic decline, poverty, and war could change societal values and debates.

      Economic growth, while not guaranteed in the future, has led to significant improvements in today's world compared to the past. However, if the future brings economic decline, poverty, and war, the responsibility for that outcome would be a topic of debate. Hawks and those concerned about fertility might gain in status, while pacifists and secular thinkers may lose. Hayek's career serves as a "white pill" as he wrote about the dangers of collectivism during the rise of Nazi Germany and Soviet Union, and although some of his views have not been proven wrong, the West has had a good run since then. The debate on the role of central planning in businesses like Amazon and Uber continues, but to me, they are an example of the market working effectively.

    • The Soviet Union's economic system was a decentralized one with government controlDespite being government-run, Soviet firms faced market tests and competition, allowing for high growth through urbanization and rebuilding, and the success of planning comes from people's dedication and the system's sustainability, not computational feasibility.

      The distinction between government and market is not as clear-cut as some argue. The Soviet Union's economic system was not purely centralized planning, but rather a decentralized one with the communist party maintaining control. The government-run firms still had to pass market tests and compete, albeit with bad incentives. This system, while not ideal, allowed for high growth during the post-war period due to urbanization and rebuilding. The market's ability to process information and allocate resources is not a result of computing a general equilibrium but rather the ability to sustain a structure that allows for survival and improvement. Hayek's argument about the computational infeasibility of central planning does not concern the speaker, as planning does not rely on solving this problem. Instead, the success of planning comes from the people involved caring about doing a good job and the system's ability to sustain itself.

    • The most intriguing aspect of competition is during market transitions from one equilibrium to another.During market transitions, producers determine what to produce and how to produce it better, leading to potential monopolies and the need for balance in decentralized systems. AI agents may create their own markets and currencies, potentially reducing transaction costs and demonstrating Hayek's theories.

      Competition, according to Herbert A. Stanley, is most interesting when markets transition from one equilibrium to another, as it's during these periods that producers figure out what to produce and how to produce it better. This perspective aligns with Peter Thiel's notion of monopolies, where competitive equilibrium leads to a lack of profits for investment in research and development or innovative projects. However, there's a potential contradiction in Hayek's call for decentralization, as it could lead to increased fragilities and the need for balance between legibility and transparency. Another insight gained from the discussion revolves around AI agents and their potential marketplaces. It's predicted that AI agents will develop their own markets and currencies, such as Bitcoin and NFTs, to facilitate transactions. This separate infrastructure could lead to a more efficient exchange of goods and services between AI agents, potentially reducing transaction costs with humans. However, it's uncertain whether this will result in a completely segregated AI economy or if it will intermix with the human economy. Regardless, this development could demonstrate Hayek's theories about the evolution of decentralized systems.

    • The Impact of AI on Personal Productivity and SocietyAI will enhance productivity for effective individuals by optimally delegating tasks, but societal implications and limitations require a nuanced approach

      As technology advances, particularly in the realm of artificial intelligence (AI), there will be a significant bifurcation in personal productivity. Effective individuals will utilize AI to optimally delegate tasks based on their comparative advantage, while less effective individuals will struggle. However, there are limitations to AI's capabilities, such as its inability to fully integrate into complex systems like general relativity and quantum mechanics. Additionally, there are uncertainties surrounding the impact of AI on society, including potential risks and ethical concerns. Hayek's prediction markets, which facilitate information aggregation through prices, could be a useful tool in navigating these uncertainties. Yet, there are challenges to implementing such markets effectively, including regulatory bottlenecks and human behavioral limitations. Overall, while AI has the potential to greatly enhance productivity, it is crucial to approach its integration into society with careful consideration and a nuanced understanding of its limitations and implications.

    • Understanding complex systems and human behavior in marketsPreserving good institutions and improving corporations requires understanding complex systems and human behavior, as some firms thrive while others don't, and resistance to change is a common challenge.

      Markets are complex systems where people solve multidimensional problems, and the price is not always a sufficient statistic. The difficulty of preserving good institutions and improving corporations over time is a puzzle, as some firms seem to get better with age, while others do not. Hayek predicted the resistance to change, which we now call nimbyism. This resistance to change is a common issue in democratic societies and even in autocratic countries. The study of economics is not only about the allocation of scarce resources but also about understanding the systems humans use, which may have more to do with universal principles than human nature alone.

    • Principles of Profit and Loss and Firm-wide DecisionsMill's ideas on profit and loss and firm-wide decision making influenced economic realities, but understanding how firms make decisions based on condensed info like P&L remains an open question. Mill's arguments on uncovering hidden structures also apply to contemporary issues like education and child rearing.

      The principles of profit and loss and selection at a firm-wide level play a significant role in shaping economic realities. However, understanding how firms disaggregate and make decisions based on condensed information like profit and loss remains an open question. Additionally, Mill's arguments about uncovering hidden structures and contingencies behind observed phenomena, such as the treatment of women throughout history, can be applied to contemporary issues like education and child rearing. Mill advocated for a balance between structured learning and free play of the imagination for children, but he would likely criticize current educational practices for their excessive coercion and wasted time. Despite his own strict upbringing, Mill's progressive ideas on child development would still challenge contemporary norms. The debate on the importance of aristocratic tutoring and selection in shaping great thinkers like Mill is ongoing, but the investment in individualized education remains a worthwhile endeavor for those who can afford it. Mill's concern for population quality and character development, as demonstrated by his own achievements, raises questions about the potential for intellectual greatness in today's world.

    • Optimistic views on societal improvement from Mill and SmithBoth Mill and Smith believed in the potential for societal improvement, recognizing changes in areas like child treatment and gender roles, despite observing limited growth. When considering AI's impact on the economy, it's essential to focus on qualitative changes and understand the underlying principles, rather than fixating on high growth rates.

      Both John Stuart Mill and Adam Smith held an optimistic view about the potential for improvement in society, despite observing limited growth and even reversals in certain areas. Smith, publishing during a period of supposed stagnation, argued for the benefits of markets, mechanization, and division of labor, extrapolating from the small growth he saw. Mill, recognizing moral progress in areas like child treatment and gender roles, believed in the possibility of further improvement. When considering the potential impact of AI on the economy, it's worth considering these thinkers' perspectives, recognizing that high, sustained growth rates may not be meaningful when the economy's composition significantly changes. While it's intriguing to speculate about the potential for 10% growth with AI, it's important to remember that such numbers may not make sense in the context of a fundamentally transformed economy. Instead, focusing on qualitative changes and understanding the principles at play may provide a more valuable perspective.

    • The Greatest Economist of All Time is a Complex QuestionThe title of the Greatest Economist of All Time is not definitively answered, as economics has seen numerous influential thinkers with unique contributions

      The title of the Greatest Economist of All Time is not a straightforward answer, as the field of economics has seen numerous influential thinkers with unique contributions. The discussion highlighted Adam Smith, but even during his time, there were notable economists like Sir James Stewart and David Hume who also made significant strides. Other economists, such as Henry George and Ronald Coase, are also deserving of recognition for their groundbreaking ideas. While some, like Thomas Schelling, have made important contributions, they may not be serious contenders for the title of GOAT. The ambiguities surrounding this question reflect the complexities of the field and the evolving nature of economic thought.

    • The Internet is transforming the way we generate and consume knowledge, with economics being a prime example.The Internet is fostering a multidisciplinary approach to economics, leading to new ideas and fields, but effective ways to subsidize and support this type of intellectual endeavor are needed.

      The way we generate and consume knowledge is undergoing a radical shift, with the Internet sphere becoming a major source of innovative ideas that transcend traditional disciplines. Economics, for instance, is no longer the sole domain of academic research, but is also being explored and advanced through Internet writing. This multidisciplinary approach, which is infectious and productive, may lead to new ideas and fields, although it might always be a minority pursuit. The Internet way of writing and thinking is a powerful new mode that is not yet fully recognized or understood, but it is shaping the way we approach complex issues, including AI and its potential impact on society. The speaker, who was influenced greatly by the history of economic thought, believes that more people, including economists and researchers, should be exposed to this mode of thinking. However, the challenge lies in finding effective ways to subsidize and support this type of intellectual endeavor.

    • Challenges of Implementing Large-Scale Subsidies into a Tax SystemThe practical challenges of implementing large-scale subsidies into a tax system for the benefit of wealth instead of GDP are significant, including the complex issue of declining interest rates and the potential for collusion in network industries, particularly in banking systems.

      While the idea of implementing large-scale subsidies into a tax system to benefit wealth instead of GDP may seem appealing, the practical challenges are significant. The trend of declining interest rates over the past century is a complex issue with no clear explanation, and the argument that network industries lead to cartel-like dynamics can be applied to social media, but not in a collusive sense. The real risk of network-based collusion lies in banking systems, where centralized institutions can facilitate collusion and exclusion from the payment system. The arguments against anarchy, while applicable to some extent to Web 3.0 crypto systems, may not fully prove that they will evolve into collusion. Instead, we can expect a continuous wave of centralization and decentralization, with natural checks embedded in the system. Anarchy exists in various forms within governments and internationally, and while some degree of collusion is necessary for things to function, the persistence of anarchy itself may lead to enough collusion to enable it to persist. Overall, it's important to acknowledge the complexities and challenges of various economic and political systems, and to continuously seek ways to improve them while accepting their inherent limitations.

    • Anarchy and the Risks of DecentralizationWhile anarchy may have historical precedents of stability, the potential risks of widespread destruction from cheap energy and advanced technology necessitate cautious decentralization and strong, benevolent governance.

      The argument against anarchy, that under anarchy anything is allowed and could lead to chaos, holds some truth. History shows us examples of anarchic societies, like medieval Iceland and Ireland, which had a degree of stability despite their lack of formal government. However, the destructiveness of anarchy is often limited by factors like low population and ineffective weapons. The speaker raises concerns about the potential for cheap energy leading to widespread destruction, and the possibility that intelligence, which has led to advancements in technology, could ultimately result in catastrophic consequences. Despite these risks, the speaker believes that the gamble of decentralization and the advantages of intelligence are worth the risk, and that we should be more consciously aware of the historical context and implications of our choices. The speaker also emphasizes the importance of strong, benevolent nations establishing dominance in emerging technologies like AI to mitigate potential risks. Ultimately, the conversation highlights the complexities and ambiguities of governance and the need for ongoing reflection and consideration of the potential consequences of our actions.

    • The potential misuse of AI by governments and ethical implicationsGovernments' role in regulating AI and ethical implications of advanced AI are complex issues requiring ongoing dialogue and careful consideration.

      The potential misuse of AI by governments, including the US, is a valid concern due to historical precedents of power corruption and human rights abuses. However, the fear of AI doom may be overblown, as some concerns, such as AI reading legal codes and threatening penalties, are not as simple or imminent as they seem. The role of governments in regulating AI labs and reacting to potential risks is uncertain, with some experts predicting inaction until a significant incident occurs. Philosophers like Robert Nozick have explored the ethical implications of advanced AI and the potential for it to treat humans as we treat animals, leading to discussions about vegetarianism and respecting the wishes of past generations. Ultimately, the complexities of AI and its potential impact on society require ongoing dialogue and careful consideration.

    • Balancing present, past, and futureWe strive for sustainability, balancing present actions with the desires of ancestors and future generations. US dollar's stability attributed to strong institutions, voter preferences, and morality. Economics trade-off: rigor vs curiosity.

      As individuals and as a society, we are faced with the challenge of balancing the desires and expectations of our ancestors and future generations with our present actions. This is a time-inconsistent problem, and there is no easy philosophical solution. Instead, we strive for sustainability and rationalize our choices based on what survives. Another intriguing topic discussed was the stability of the US dollar and the well-managed monetary policies in the US. The reasons for this are complex, with factors including voter preferences, wealth, and the reputation and incentives of those in charge of these institutions. However, it's important to note that not all unelected bodies are well-run, and there are plenty of corrupt ones around the world. The high standards and strong sense of morality and self-interest among those in the US institutions, particularly the Federal Reserve, contribute to their effectiveness. Regarding economics, there's been a shift towards specialization and rigor, but at the cost of curiosity and a broad perspective. This trade-off, while understandable, can limit the field's ability to tackle the biggest questions. In summary, navigating the complexities of time inconsistency, monetary policies, and the evolution of economics requires a thoughtful and nuanced approach.

    • The sum of economic theories is greater than the wholeIndividual economic theories may not lead to a grand understanding, but their application can create larger, interesting ideas and generate income.

      While the economic theories of figures like Jevons contributed significantly to the field, the proliferation of specialists and focus on micro work may not lead to a grand theory or cohesive understanding in academic disciplines. However, the sum of these parts can be greater than the whole when individuals like Tyler Cowen apply this knowledge to create larger, interesting ideas and make a living from it. The future of this thinking may involve adapting to the rise of AI and finding ways to monetize audiences. Regarding criticisms of democratic capitalist models, the libertarian critique of a drift towards socialism may not be validated by the current state of countries like France and Sweden. The egalitarian critique of income inequality and its impact on political and moral equality is also debatable, as countries with high inequality, like the US, can still function relatively well. The Nietzschean critique, which warns of the potential end of civilization, is a complex issue, and it's unclear if the market is the primary threat. Overall, the challenges facing democratic capitalist models are manageable compared to those of previous eras.

    • Investing in resources and talent to tackle challenges despite existential risksDespite potential risks, we should remain optimistic and invest in resources and talent to address challenges, including nuclear proliferation, AI limits, and geopolitical risks. Productivity may increase due to the increasing number of researchers and rising wages.

      Despite the potential existential risks, such as nuclear proliferation and the limits of AI, we should remain optimistic and continue to invest in resources and talent to tackle these challenges. The market seems to be rationalizing geopolitical risks, but volatility, as measured by the VIX index, may increase at some point. The conversation also touched upon the limitations of scaling and the frontier of scientific discovery, but even if researcher productivity may be declining, the increasing number of researchers and rising wages could lead to productivity gains. While AI is expected to raise productivity growth, it may not be as revolutionary as some believe. The conversation also highlighted the importance of measuring productivity in terms of wages and the potential impact of research and development in various industries.

    • The impact of AI might not be linearly related to resources, but rather the quality of integrationThe progress of AI might not be as extraordinary as imagined, but the integration of high-quality systems can lead to significant advancements.

      The progress of AI development might not be linearly related to the number of copies or resources we have, but rather, the quality and integration of the best AI system into our existing systems. The impact of AI could be compared to the disproportionate influence of a small population of Jews in the 20th century, leading to significant advancements. However, the pace of change might not be as extraordinary as some might imagine, and we should expect a mix of positive outcomes and disruptions. The speaker's perspective is influenced by his belief in a decentralized world and the challenges of managing AI progress. He is currently writing a book, "The Marginal Revolution," focusing on the actual margin and the importance of context. The speaker encourages listeners to engage with the podcast and share it with others.

    Recent Episodes from Dwarkesh Podcast

    Tony Blair - Life of a PM, The Deep State, Lee Kuan Yew, & AI's 1914 Moment

    Tony Blair - Life of a PM, The Deep State, Lee Kuan Yew, & AI's 1914 Moment

    I chatted with Tony Blair about:

    - What he learned from Lee Kuan Yew

    - Intelligence agencies track record on Iraq & Ukraine

    - What he tells the dozens of world leaders who come seek advice from him

    - How much of a PM’s time is actually spent governing

    - What will AI’s July 1914 moment look like from inside the Cabinet?

    Enjoy!

    Watch the video on YouTube. Read the full transcript here.

    Follow me on Twitter for updates on future episodes.

    Sponsors

    - Prelude Security is the world’s leading cyber threat management automation platform. Prelude Detect quickly transforms threat intelligence into validated protections so organizations can know with certainty that their defenses will protect them against the latest threats. Prelude is backed by Sequoia Capital, Insight Partners, The MITRE Corporation, CrowdStrike, and other leading investors. Learn more here.

    - This episode is brought to you by Stripe, financial infrastructure for the internet. Millions of companies from Anthropic to Amazon use Stripe to accept payments, automate financial processes and grow their revenue.

    If you’re interested in advertising on the podcast, check out this page.

    Timestamps

    (00:00:00) – A prime minister’s constraints

    (00:04:12) – CEOs vs. politicians

    (00:10:31) – COVID, AI, & how government deals with crisis

    (00:21:24) – Learning from Lee Kuan Yew

    (00:27:37) – Foreign policy & intelligence

    (00:31:12) – How much leadership actually matters

    (00:35:34) – Private vs. public tech

    (00:39:14) – Advising global leaders

    (00:46:45) – The unipolar moment in the 90s



    Get full access to Dwarkesh Podcast at www.dwarkeshpatel.com/subscribe
    Dwarkesh Podcast
    enJune 26, 2024

    Francois Chollet, Mike Knoop - LLMs won’t lead to AGI - $1,000,000 Prize to find true solution

    Francois Chollet, Mike Knoop - LLMs won’t lead to AGI - $1,000,000 Prize to find true solution

    Here is my conversation with Francois Chollet and Mike Knoop on the $1 million ARC-AGI Prize they're launching today.

    I did a bunch of socratic grilling throughout, but Francois’s arguments about why LLMs won’t lead to AGI are very interesting and worth thinking through.

    It was really fun discussing/debating the cruxes. Enjoy!

    Watch on YouTube. Listen on Apple PodcastsSpotify, or any other podcast platform. Read the full transcript here.

    Timestamps

    (00:00:00) – The ARC benchmark

    (00:11:10) – Why LLMs struggle with ARC

    (00:19:00) – Skill vs intelligence

    (00:27:55) - Do we need “AGI” to automate most jobs?

    (00:48:28) – Future of AI progress: deep learning + program synthesis

    (01:00:40) – How Mike Knoop got nerd-sniped by ARC

    (01:08:37) – Million $ ARC Prize

    (01:10:33) – Resisting benchmark saturation

    (01:18:08) – ARC scores on frontier vs open source models

    (01:26:19) – Possible solutions to ARC Prize



    Get full access to Dwarkesh Podcast at www.dwarkeshpatel.com/subscribe
    Dwarkesh Podcast
    enJune 11, 2024

    Leopold Aschenbrenner - China/US Super Intelligence Race, 2027 AGI, & The Return of History

    Leopold Aschenbrenner - China/US Super Intelligence Race, 2027 AGI, & The Return of History

    Chatted with my friend Leopold Aschenbrenner on the trillion dollar nationalized cluster, CCP espionage at AI labs, how unhobblings and scaling can lead to 2027 AGI, dangers of outsourcing clusters to Middle East, leaving OpenAI, and situational awareness.

    Watch on YouTube. Listen on Apple PodcastsSpotify, or any other podcast platform. Read the full transcript here.

    Follow me on Twitter for updates on future episodes. Follow Leopold on Twitter.

    Timestamps

    (00:00:00) – The trillion-dollar cluster and unhobbling

    (00:20:31) – AI 2028: The return of history

    (00:40:26) – Espionage & American AI superiority

    (01:08:20) – Geopolitical implications of AI

    (01:31:23) – State-led vs. private-led AI

    (02:12:23) – Becoming Valedictorian of Columbia at 19

    (02:30:35) – What happened at OpenAI

    (02:45:11) – Accelerating AI research progress

    (03:25:58) – Alignment

    (03:41:26) – On Germany, and understanding foreign perspectives

    (03:57:04) – Dwarkesh’s immigration story and path to the podcast

    (04:07:58) – Launching an AGI hedge fund

    (04:19:14) – Lessons from WWII

    (04:29:08) – Coda: Frederick the Great



    Get full access to Dwarkesh Podcast at www.dwarkeshpatel.com/subscribe
    Dwarkesh Podcast
    enJune 04, 2024

    John Schulman (OpenAI Cofounder) - Reasoning, RLHF, & Plan for 2027 AGI

    John Schulman (OpenAI Cofounder) - Reasoning, RLHF, & Plan for 2027 AGI

    Chatted with John Schulman (cofounded OpenAI and led ChatGPT creation) on how posttraining tames the shoggoth, and the nature of the progress to come...

    Watch on YouTube. Listen on Apple PodcastsSpotify, or any other podcast platform. Read the full transcript here. Follow me on Twitter for updates on future episodes.

    Timestamps

    (00:00:00) - Pre-training, post-training, and future capabilities

    (00:16:57) - Plan for AGI 2025

    (00:29:19) - Teaching models to reason

    (00:40:50) - The Road to ChatGPT

    (00:52:13) - What makes for a good RL researcher?

    (01:00:58) - Keeping humans in the loop

    (01:15:15) - State of research, plateaus, and moats

    Sponsors

    If you’re interested in advertising on the podcast, fill out this form.

    * Your DNA shapes everything about you. Want to know how? Take 10% off our Premium DNA kit with code DWARKESH at mynucleus.com.

    * CommandBar is an AI user assistant that any software product can embed to non-annoyingly assist, support, and unleash their users. Used by forward-thinking CX, product, growth, and marketing teams. Learn more at commandbar.com.



    Get full access to Dwarkesh Podcast at www.dwarkeshpatel.com/subscribe
    Dwarkesh Podcast
    enMay 15, 2024

    Mark Zuckerberg - Llama 3, Open Sourcing $10b Models, & Caesar Augustus

    Mark Zuckerberg - Llama 3, Open Sourcing $10b Models, & Caesar Augustus

    Mark Zuckerberg on:

    - Llama 3

    - open sourcing towards AGI

    - custom silicon, synthetic data, & energy constraints on scaling

    - Caesar Augustus, intelligence explosion, bioweapons, $10b models, & much more

    Enjoy!

    Watch on YouTube. Listen on Apple PodcastsSpotify, or any other podcast platform. Human edited transcript with helpful links here.

    Timestamps

    (00:00:00) - Llama 3

    (00:08:32) - Coding on path to AGI

    (00:25:24) - Energy bottlenecks

    (00:33:20) - Is AI the most important technology ever?

    (00:37:21) - Dangers of open source

    (00:53:57) - Caesar Augustus and metaverse

    (01:04:53) - Open sourcing the $10b model & custom silicon

    (01:15:19) - Zuck as CEO of Google+

    Sponsors

    If you’re interested in advertising on the podcast, fill out this form.

    * This episode is brought to you by Stripe, financial infrastructure for the internet. Millions of companies from Anthropic to Amazon use Stripe to accept payments, automate financial processes and grow their revenue. Learn more at stripe.com.

    * V7 Go is a tool to automate multimodal tasks using GenAI, reliably and at scale. Use code DWARKESH20 for 20% off on the pro plan. Learn more here.

    * CommandBar is an AI user assistant that any software product can embed to non-annoyingly assist, support, and unleash their users. Used by forward-thinking CX, product, growth, and marketing teams. Learn more at commandbar.com.



    Get full access to Dwarkesh Podcast at www.dwarkeshpatel.com/subscribe

    Sholto Douglas & Trenton Bricken - How to Build & Understand GPT-7's Mind

    Sholto Douglas & Trenton Bricken - How to Build & Understand GPT-7's Mind

    Had so much fun chatting with my good friends Trenton Bricken and Sholto Douglas on the podcast.

    No way to summarize it, except: 

    This is the best context dump out there on how LLMs are trained, what capabilities they're likely to soon have, and what exactly is going on inside them.

    You would be shocked how much of what I know about this field, I've learned just from talking with them.

    To the extent that you've enjoyed my other AI interviews, now you know why.

    So excited to put this out. Enjoy! I certainly did :)

    Watch on YouTube. Listen on Apple PodcastsSpotify, or any other podcast platform. 

    There's a transcript with links to all the papers the boys were throwing down - may help you follow along.

    Follow Trenton and Sholto on Twitter.

    Timestamps

    (00:00:00) - Long contexts

    (00:16:12) - Intelligence is just associations

    (00:32:35) - Intelligence explosion & great researchers

    (01:06:52) - Superposition & secret communication

    (01:22:34) - Agents & true reasoning

    (01:34:40) - How Sholto & Trenton got into AI research

    (02:07:16) - Are feature spaces the wrong way to think about intelligence?

    (02:21:12) - Will interp actually work on superhuman models

    (02:45:05) - Sholto’s technical challenge for the audience

    (03:03:57) - Rapid fire



    Get full access to Dwarkesh Podcast at www.dwarkeshpatel.com/subscribe

    Demis Hassabis - Scaling, Superhuman AIs, AlphaZero atop LLMs, Rogue Nations Threat

    Demis Hassabis - Scaling, Superhuman AIs, AlphaZero atop LLMs, Rogue Nations Threat

    Here is my episode with Demis Hassabis, CEO of Google DeepMind

    We discuss:

    * Why scaling is an artform

    * Adding search, planning, & AlphaZero type training atop LLMs

    * Making sure rogue nations can't steal weights

    * The right way to align superhuman AIs and do an intelligence explosion

    Watch on YouTube. Listen on Apple PodcastsSpotify, or any other podcast platform. Read the full transcript here.

    Timestamps

    (0:00:00) - Nature of intelligence

    (0:05:56) - RL atop LLMs

    (0:16:31) - Scaling and alignment

    (0:24:13) - Timelines and intelligence explosion

    (0:28:42) - Gemini training

    (0:35:30) - Governance of superhuman AIs

    (0:40:42) - Safety, open source, and security of weights

    (0:47:00) - Multimodal and further progress

    (0:54:18) - Inside Google DeepMind



    Get full access to Dwarkesh Podcast at www.dwarkeshpatel.com/subscribe

    Patrick Collison (Stripe CEO) - Craft, Beauty, & The Future of Payments

    Patrick Collison (Stripe CEO) - Craft, Beauty, & The Future of Payments

    We discuss:

    * what it takes to process $1 trillion/year

    * how to build multi-decade APIs, companies, and relationships

    * what's next for Stripe (increasing the GDP of the internet is quite an open ended prompt, and the Collison brothers are just getting started).

    Plus the amazing stuff they're doing at Arc Institute, the financial infrastructure for AI agents, playing devil's advocate against progress studies, and much more.

    Watch on YouTube. Listen on Apple PodcastsSpotify, or any other podcast platform. Read the full transcript here. Follow me on Twitter for updates on future episodes.

    Timestamps

    (00:00:00) - Advice for 20-30 year olds

    (00:12:12) - Progress studies

    (00:22:21) - Arc Institute

    (00:34:27) - AI & Fast Grants

    (00:43:46) - Stripe history

    (00:55:44) - Stripe Climate

    (01:01:39) - Beauty & APIs

    (01:11:51) - Financial innards

    (01:28:16) - Stripe culture & future

    (01:41:56) - Virtues of big businesses

    (01:51:41) - John



    Get full access to Dwarkesh Podcast at www.dwarkeshpatel.com/subscribe

    Tyler Cowen - Hayek, Keynes, & Smith on AI, Animal Spirits, Anarchy, & Growth

    Tyler Cowen - Hayek, Keynes, & Smith on AI, Animal Spirits, Anarchy, & Growth

    It was a great pleasure speaking with Tyler Cowen for the 3rd time.

    We discussed GOAT: Who is the Greatest Economist of all Time and Why Does it Matter?, especially in the context of how the insights of Hayek, Keynes, Smith, and other great economists help us make sense of AI, growth, animal spirits, prediction markets, alignment, central planning, and much more.

    The topics covered in this episode are too many to summarize. Hope you enjoy!

    Watch on YouTube. Listen on Apple PodcastsSpotify, or any other podcast platform. Read the full transcript here. Follow me on Twitter for updates on future episodes.

    Timestamps

    (0:00:00) - John Maynard Keynes

    (00:17:16) - Controversy

    (00:25:02) - Fredrick von Hayek

    (00:47:41) - John Stuart Mill

    (00:52:41) - Adam Smith

    (00:58:31) - Coase, Schelling, & George

    (01:08:07) - Anarchy

    (01:13:16) - Cheap WMDs

    (01:23:18) - Technocracy & political philosophy

    (01:34:16) - AI & Scaling



    Get full access to Dwarkesh Podcast at www.dwarkeshpatel.com/subscribe

    Lessons from The Years of Lyndon Johnson by Robert Caro [Narration]

    Lessons from The Years of Lyndon Johnson by Robert Caro [Narration]

    This is a narration of my blog post, Lessons from The Years of Lyndon Johnson by Robert Caro.

    You read the full post here: https://www.dwarkeshpatel.com/p/lyndon-johnson

    Listen on Apple Podcasts, Spotify, or any other podcast platform. Follow me on Twitter for updates on future posts and episodes.



    Get full access to Dwarkesh Podcast at www.dwarkeshpatel.com/subscribe

    Related Episodes

    Finding Yourself To Become A Better Leader with Wade Eyerly

    Finding Yourself To Become A Better Leader with Wade Eyerly

    Finding Yourself To Become A Better Leader with Wade Eyerly - Insurance Leader EP25

    On this episode, Daniel and Ryan discuss all sorts of topics with Wade Eyerly, from finding yourself in order to develop into a better leader, starting a private jet charter airline, working at the White House, and more. Wade is an incredible leader with vast experience as a CEO, investment banker, author, innovator, and leader.

    Learn More About Wade Eyerly >>>



    To contact Daniel, email him at Daniel@GreatParkInsurance.com


    To contact Ryan, email him at Rreynolds@GrimesInsurance.com

    Vitaliy Katsenelson: How To Boost IQ and Transform Your Brain

    Vitaliy Katsenelson: How To Boost IQ and Transform Your Brain

    Can you increase your IQ in just 15 minutes a day?

    How easy is it to transform your brain to become more creative, intelligent and productive?

    With AI positioned to take over our jobs and our hobbies, we have to pause and consider the true purpose of writing, art and music.

    And I'm honored that joining us here today is Vitaliy Katsenelson, an award-winning writer, the CEO of Investment Management Associates, and author of a number of books, including his latest, “Soul in the Game”.

    Vitaliy is a brilliant writer, multifaceted intellectual and an incredible person, so I hope you enjoy this conversation. We're chatting about:

    • How to boost IQ in as little as 15 minutes a day while transforming your brain into a writer's brain
    • The dark side of over-abundance and the value of self-imposed constraints
    • Why fewer choices often lead to more happiness
    • Why an investment guy is telling us how unimportant money is
    • How art trains us to be mindful and look for beauty in the world around us
    • And tons more…

    Read the show notes: https://fatburningman.com/vitaliy-katsenelson-how-to-boost-iq-and-transform-your-brain/ 

    Go to Investor.fm to read Vitaliy’s articles, subscribe to The Intellectual Investor podcast, get books and much more.

    Join The Wild Guild and get freebies on Patreon: https://www.patreon.com/abeljames 

    Watch on YouTube: https://www.youtube.com/fatburningman 

    Like the show on Facebook: http://www.facebook.com/fatburningman 

    Follow on Twitter: http://www.twitter.com/fatburnman 

    Click here for your free Fat-Burning Kit: http://fatburningman.com/bonus 

    Sponsor Spotlight: ft. Alumni Ventures’ CEO Mike Collins | How to invest in startups as individual investors?

    Sponsor Spotlight: ft. Alumni Ventures’ CEO Mike Collins | How to invest in startups as individual investors?

    Sponsor Spotlight guest: Alumni Ventures’ CEO Mike Collins 

    Welcome to the Smart Venture Podcast’s sponsor spotlight, where we take a moment to highlight the companies making our show possible. We're thrilled to feature Alumni Ventures.  Alumni Ventures, offers individual investors access to venture investing through its diversified, professional-grade venture portfolios. We're excited to share more about what they do and how they're making a difference. So sit back, relax, and let's dive into Alumni Venture’s sponsor spotlight. 

    _________________________________________________________________ 

    How to invest in startups as individual investors?

    I get this question, A LOT.

    So I brought one of the experts to join me to talk about this.

    Mike Collins is the CEO of Alumni Ventures, America's Largest Venture Capital Firm for Individual Investors.

    Smart Venture Podcast’s sponsor, Alumni Ventures, offers individual investors access to venture investing through its diversified, professional-grade venture portfolios. The company’s funds have consistently outperformed public market equivalents, with over $1.1 billion raised and invested in over eleven hundred portfolio companies. They have a dedicated team of fifty full-time venture investors and were the #1 most active venture firm in the US in 2022 and #3 globally according to Pitchbook. Investing in venture capital can help reduce overall portfolio risk and increase the likelihood of stronger returns. To learn more, visit av.vc/grace and schedule a call. Investors must be accredited. Please note that all financial investments involve risk, past performance does not guarantee future results, and it is important to conduct your own research and seek professional advice before making any investment decisions.

    Check out their service: av.vc/grace

    =====================

    =====================

     

    Join the SVP fam with your host Grace Gong. In each episode, we are going to have conversations with some of the top investors, superstar founders, as well as well-known tech executives in silicon valley. We will have a coffee chat with them to learn their ways of thinking and actionable tips on how to build or invest in a successful company.

    Nvidia Part III: The Dawn of the AI Era (2022-2023)

    Nvidia Part III: The Dawn of the AI Era (2022-2023)

    It’s a(nother) new era for Nvidia.

    We thought we’d closed the Acquired book on Nvidia back in April 2022. The story was all wrapped up: Jensen & crew had set out on an amazing journey to accelerate the world’s computing workloads. Along the way they’d discovered a wondrous opportunity (machine learning powered social media feed recommendations). They forged incredible Power in the CUDA platform, and used it to triumph over seemingly insurmountable adversity — the stock market penalty-box.

    But, it turned out that was only the precursor to an even wilder journey. Over the past 18 months Nvidia has weathered one of the steepest stock crashes in history ($500B+ market cap wiped away peak-to-trough!). And, it has of course also experienced an even more fantastical rise — becoming the platform that’s powering the emergence of perhaps a new form of intelligence itself… and in the process becoming a trillion-dollar company.

    Today we tell another chapter in the amazing Nvidia saga: the dawn of the AI era. Tune in!

    Sponsors:

    Thanks to our fantastic partners, any member of the Acquired community can now get:

    More Acquired!:

    Links:

    Carve Outs:

    ‍Note: Acquired hosts and guests may hold assets discussed in this episode. This podcast is not investment advice, and is intended for informational and entertainment purposes only. You should do your own research and make your own independent decisions when considering any financial transactions.