reSee.it - Related Video Feed

The speaker discusses the need to create records for displaced people, such as refugees, to help them interact with the world economy and move between places while keeping important information secure. They mention the importance of medical records and vaccination history, particularly for travel purposes. The speaker also mentions the use of blockchain technology to link physical documents to a controlled record-keeping system, ensuring tamper-proof authenticity. They highlight the significance of provenance and counterfeiting prevention in this context.

Speaker 0 expresses a vision to transform government transparency and control over spending. The core goal is to blockchain the entire federal government, and to have every dime of federal spending online in real time, so there is day-by-day, month-by-month visibility into what the Department of Interior and Veterans Affairs (and other agencies) are spending money on. The speaker suggests there could be national security risks with such transparency, noting that some aspects could be “black box” or restricted, but asserts the ideal is real-time visibility into government spending. The speaker argues that the public should know exactly how money is spent, asking concrete questions like what the Department of Interior is spending money on, and whether they are buying items such as “$50 hammers” or “$200 bandages.” The overarching point is that this is “our money” and “we are the sovereign,” because “we create the government,” we earn the money, and “the government extract it from us with our consent.” Therefore, there is a right to know where the money goes.

Speaker 0 expresses opposition to cryptocurrency. Speaker 1 mentions that Jamie, who supports blockchain, helped launch JPMorganCoin. They explain that JPMorgan created its own blockchain protocol based on Ethereum, allowing private transactions. Speaker 0 suggests that the only use case for blockchain is criminal activity. Speaker 2 states that JPMorgan was involved in Ethereum from the beginning and played a major role in the Enterprise Ethereum Alliance. Speaker 0 comments on shutting down blockchain if they were the government. Speaker 3 compares the Mt. Gox scandal to Bernie Madoff's Ponzi scheme, where JPMorgan was involved. JPMorgan account holders sued the bank and recovered over $2 billion, but no executives went to jail.

Speaker 0 mentions that adopting the technology is not the first thing they do, but rather the last. Speaker 1 discusses Ripple, a company known for being a leader in Enterprise blockchain. They mention that Ripple holds a significant amount of a cryptocurrency they created, but it hasn't gained much adoption. Despite this, the company is becoming wealthy. The speaker wonders if Ripple can make the cryptocurrency live up to its value.

A discussion centers on a new proposed law, HR 8250, which would require operating system providers to verify the age of any user of an operating system and for other purposes, covering Windows, macOS, iOS, and Android, with open-source Linux considered in the debate. The claim is that this could serve as a Trojan horse to control people through a digital ID system, rather than being merely safety-focused. Speaker 1 references Catherine Austin Fitz, who says that if global elites deploy digital ID systems, they will control all aspects, including health freedoms and financial transactions. She argues that once financial transaction control is in place, all protections in health and food freedoms could be negated, and a 100% digital system with a digital ID and programmable money would allow authorities to dictate health decisions, vaccine status, gender-transition decisions for children, and other policies by turning off funds. Speaker 0 notes that Fitz is not hyperbolic and mentions Austin Steinbart, founder of the Quantum Party of America, who is joined by Speaker 0 to discuss the issue further. Speaker 2 (Austin Steinbart) asserts that the HR 8250 proposal is a disaster and goes beyond a digital ID concept by embedding age verification into the core of every device. He says the bill is six pages long and delegates enforcement to the FTC, creating ambiguity about whether biometrics, ID cards, or face scans would be used, leaving the mechanism up to the executive branch. He points out that the proposal could coordinate with companies like Apple (potentially via Face ID) and Microsoft to embed verification, while raising questions about how open-source Linux distributions would be forced to comply. He notes that Linux is open-source and typically users have root access, enabling workarounds or removal of such core files, and questions how a retrospective integration would work on devices like POS systems or hotel front-desk computers. Speaker 0 asks how the implementation would occur and whether the digital ID is the core objective beyond age verification. Speaker 2 confirms that the core goal is a universal digital ID across platforms, tying to privacy and cybersecurity concerns by requiring every service to interact with core OS files to verify age, with California already moving toward age verification that apps and websites would rely on. Speaker 0 links this to a broader move toward a central bank digital currency (CBDC) and a digital ID, quoting a sound bite from Catherine Austin Fitz about health identifiers affecting travel and other activities. Speaker 3 (a figure from the World Economic Forum) is cited, emphasizing tokenization of financial assets and the rapid rollout of a digital wallet and digitized currencies globally, with a critique that many countries are unprepared for such changes. Speaker 2 clarifies that blockchain or tokenization per se isn’t inherently bad, but concerns arise when centralized actors with anti-freedom aims design and control the system, shaping speech and policy. They discuss the potential benefits of tokenized assets in theory, while warning that centralized control could enable censorship and restricted financial activity. Speaker 0 ends by urging viewers to contact members of Congress to oppose HR 8250, urging them to “burn this thing down,” and thanks Speaker 2 for the analysis.

Donald Trump suggested Elon Musk audit the federal government. One speaker believes AI can democratize government and increase transparency, or enslave citizens to the government and intelligence agencies, and that Musk understands this best. The Pentagon has failed every audit for the last 20 years and lost $4.3 trillion in the last audit. This money was primarily lost on equipment purchases whose locations are unknown, forcing the Pentagon to repurchase them. These problems are solvable with AI, which could track stockpiles and warehouses to identify the location of equipment.

Donald Trump suggested Elon Musk audit the federal government. One speaker believes AI can democratize government and increase transparency, but could also enslave citizens to the government and intelligence agencies. They feel Musk understands this best and could deploy AI effectively. The Pentagon has failed every audit for the last 20 years and lost $4.3 trillion in the last audit, mostly due to being unable to locate purchased equipment. AI could fix these problems by tracking stockpiles and warehouses to eliminate government waste.

The speaker discusses the potential risks of decentralized technology. They explain that while this technology aims to liberate individuals and give them control over their personal information, it could also be used by centralized powers to trap people. The speaker suggests that governments could create their own centralized blockchain, turning their currency into a permissioned cryptocurrency. This would allow governments to have complete knowledge of individuals' transactions and diminish privacy. They mention the example of the Marshall Islands, which passed a law for a decentralized currency but most governments may not be willing to give up control over their monetary policy. The interviewer mentions that Joseph Flubin, a co-founder of Ethereum, disagrees with this pessimistic view, considering it fear, uncertainty, and doubt (FUD).

The speaker discusses the battle between crypto and the government, particularly the SEC. They explain that the US government is interested in slowing or killing crypto due to their preference for intermediaries and centralized control. However, they believe that the ecosystem can continue to operate globally and in the US with more focus on decentralization. They mention that the Ripple XRP ruling was favorable to centralized exchanges and wallets. The speaker also talks about the clash between centralized and decentralized trust and the need for both to coexist. They advocate for regulating use cases rather than stifling tech innovation.

Brad made a compelling presentation to the commission and FTC, emphasizing the need for decentralization of the web. He highlighted the growing consolidation of power among Google, Apple, Facebook, and Amazon, making it impossible to compete by simply offering a better service. Drawing parallels to the mid-nineties when Microsoft dominated the PC software industry, Brad explained that a shift in venue to the web and a change in business model to open source ultimately disrupted Microsoft's hegemony. To compete with dominant data monopolies, such as Google and Facebook, the game needs to be changed. Blockchains, as open public data stores, offer the best chance for innovation and bottom-up startup growth. This argument was presented to the SEC as the next wave of technology.

- The conversation centers on how AI progress has evolved over the last few years, what is surprising, and what the near future might look like in terms of capabilities, diffusion, and economic impact. - Big picture of progress - Speaker 1 argues that the underlying exponential progression of AI tech has followed expectations, with models advancing from “smart high school student” to “smart college student” to capabilities approaching PhD/professional levels, and code-related tasks extending beyond that frontier. The pace is roughly as anticipated, with some variance in direction for specific tasks. - The most surprising aspect, per Speaker 1, is the lack of public recognition of how close we are to the end of the exponential growth curve. He notes that public discourse remains focused on political controversies while the technology is approaching a phase where the exponential growth tapers or ends. - What “the exponential” looks like now - There is a shared hypothesis dating back to 2017 (the big blob of compute hypothesis) that what matters most for progress are a small handful of factors: compute, data quantity, data quality/distribution, training duration, scalable objective functions, and normalization/conditioning for stability. - Pretraining scaling has continued to yield gains, and now RL shows a similar pattern: pretraining followed by RL phases can scale with long-term training data and objectives. Tasks like math contests have shown log-linear improvements with training time in RL, and this pattern mirrors pretraining. - The discussion emphasizes that RL and pretraining are not fundamentally different in their relation to scaling; RL is seen as an RL-like extension atop the same scaling principles already observed in pretraining. - On the nature of learning and generalization - There is debate about whether the best path to generalization is “human-like” learning (continual on-the-job learning) or large-scale pretraining plus RL. Speaker 1 argues the generalization observed in pretraining on massive, diverse data (e.g., Common Crawl) is what enables the broad capabilities, and RL similarly benefits from broad, varied data and tasks. - The in-context learning capacity is described as a form of short- to mid-term learning that sits between long-term human learning and evolution, suggesting a spectrum rather than a binary gap between AI learning and human learning. - On the end state and timeline to AGI-like capabilities - Speaker 1 expresses high confidence (~90% or higher) that within ten years we will reach capabilities where a country-of-geniuses-level model in a data center could handle end-to-end tasks (including coding) and generalize across many domains. He places a strong emphasis on timing: “one to three years” for on-the-job, end-to-end coding and related tasks; “three to five” or “five to ten” years for broader, high-ability AI integration into real work. - A central caution is the diffusion problem: even if the technology is advancing rapidly, the economic uptake and deployment into real-world tasks take time due to organizational, regulatory, and operational frictions. He envisions two overlapping fast exponential curves: one for model capability and one for diffusion into the economy, with the latter slower but still rapid compared with historical tech diffusion. - On coding and software engineering - The conversation explores whether the near-term future could see 90% or even 100% of coding tasks done by AI. Speaker 1 clarifies his forecast as a spectrum: - 90% of code written by models is already seen in some places. - 90% of end-to-end SWE tasks (including environment setup, testing, deployment, and even writing memos) might be handled by models; 100% is still a broader claim. - The distinction is between what can be automated now and the broader productivity impact across teams. Even with high automation, human roles in software design and project management may shift rather than disappear. - The value of coding-specific products like Claude Code is discussed as a result of internal experimentation becoming externally marketable; adoption is rapid in the coding domain, both internally and externally. - On product strategy and economics - The economics of frontier AI are discussed in depth. The industry is characterized as a few large players with steep compute needs and a dynamic where training costs grow rapidly while inference margins are substantial. This creates a cycle: training costs are enormous, but inference revenue plus margins can be significant; the industry’s profitability depends on accurately forecasting future demand for compute and managing investment in training versus inference. - The concept of a “country of geniuses in a data center” is used to describe the point at which frontier AI capabilities become so powerful that they unlock large-scale economic value. The timing is uncertain and depends on both technical progress and the diffusion of benefits through the economy. - There is a nuanced view on profitability: in a multi-firm equilibrium, each model may be profitable on its own, but the cost of training new models can outpace current profits if demand does not grow as fast as the compute investments. The balance is described in terms of a distribution where roughly half of compute is used for training and half for inference, with margins on inference driving profitability while training remains a cost center. - On governance, safety, and society - The conversation ventures into governance and international dynamics. The world may evolve toward an “AI governance architecture” with preemption or standard-setting at the federal level, to avoid an unhelpful patchwork of state laws. The idea is to establish standards for transparency, safety, and alignment while balancing innovation. - There is concern about autocracies and the potential for AI to exacerbate geopolitical tensions. The idea is that the post-AGI world may require new governance structures that preserve human freedoms, while enabling competitive but safe AI development. Speaker 1 contemplates scenarios in which authoritarian regimes could become destabilized by powerful AI-enabled information and privacy tools, though cautions that practical governance approaches would be required. - The role of philanthropy is acknowledged, but there is emphasis on endogenous growth and the dissemination of benefits globally. Building AI-enabled health, drug discovery, and other critical sectors in the developing world is seen as essential for broad distribution of AI benefits. - The role of safety tools and alignments - Anthropic’s approach to model governance includes a constitution-like framework for AI behavior, focusing on principles rather than just prohibitions. The idea is to train models to act according to high-level principles with guardrails, enabling better handling of edge cases and greater alignment with human values. - The constitution is viewed as an evolving set of guidelines that can be iterated within the company, compared across different organizations, and subject to broader societal input. This iterative approach is intended to improve alignment while preserving safety and corrigibility. - Specific topics and examples - Video editing and content workflows illustrate how an AI with long-context capabilities and computer-use ability could perform complex tasks, such as reviewing interviews, identifying where to edit, and generating a final cut with context-aware decisions. - There is a discussion of long-context capacity (from thousands of tokens to potentially millions) and the engineering challenges of serving such long contexts, including memory management and inference efficiency. The conversation stresses that these are engineering problems tied to system design rather than fundamental limits of the model’s capabilities. - Final outlook and strategy - The timeline for a country-of-geniuses in a data center is framed as potentially within one to three years for end-to-end on-the-job capabilities, and by 2028-2030 for broader societal diffusion and economic impact. The probability of reaching fundamental capabilities that enable trillions of dollars in revenue is asserted as high within the next decade, with 2030 as a plausible horizon. - There is ongoing emphasis on responsible scaling: the pace of compute expansion must be balanced with thoughtful investment and risk management to ensure long-term stability and safety. The broader vision includes global distribution of benefits, governance mechanisms that preserve civil liberties, and a cautious but optimistic expectation that AI progress will transform many sectors while requiring careful policy and institutional responses. - Mentions of concrete topics - Claude Code as a notable Anthropic product rising from internal use to external adoption. - The idea of a “collective intelligence” approach to shaping AI constitutions with input from multiple stakeholders, including potential future government-level processes. - The role of continual learning, model governance, and the interplay between technology progression and regulatory development. - The broader existential and geopolitical questions—how the world navigates diffusion, governance, and potential misalignment—are acknowledged as central to both policy and industry strategy. - In sum, the dialogue canvasses (a) the expected trajectory of AI progress and the surprising proximity to exponential endpoints, (b) how scaling, pretraining, and RL interact to yield generalization, (c) the practical timelines for on-the-job competencies and automation of complex professional tasks, (d) the economics of compute and the diffusion of frontier AI across the economy, (e) governance, safety, and the potential for a governance architecture (constitutions, preemption, and multi-stakeholder input), and (f) the strategic moves of Anthropic (including Claude Code) within this evolving landscape.

In this video, the speaker, who has a finance background, discusses the importance of combining computer science, markets, law, and social norms when it comes to blockchain technology. They mention the lack of a widespread enterprise solution for blockchain-based payments and emphasize the significance of public policy. The speaker also mentions their own stance on Bitcoin and smart contracts, stating that they lean towards being a minimalist. They mention another person, Larry, who may have a more maximalist view. The speaker acknowledges that their perspective may change based on the discussions and teachings in the class. They briefly touch on permissioned and permissionless blockchain systems and mention a startup that is working on a permission system.