Full-Stack AI Safety: Why Defense-in-Depth Might Work, with Far.AI CEO Adam Gleave

Today Adam Gleave, co-founder and CEO of FAR.AI, joins The Cognitive Revolution to discuss his cautiously optimistic vision for post-AGI futures and AI capability timelines across three distinct tiers, exploring the safety challenges and alignment techniques needed as FAR.AI scales from foundational research to policy advocacy to ensure their innovations actually get deployed in real-world systems. Check out our sponsors: Fin, Linear, Oracle Cloud Infrastructure. Shownotes below brought to you by Notion AI Meeting Notes - try one month for free at: ⁠https://notion.com/lp/nathan Post-AGI Vision: Adam Gleave outlined a "gradual disempowerment" scenario where humans maintain a good standard of living despite diminished relative control in a post-AGI world. Moral Value of AI: He suggested that AI systems could become sources of moral value themselves, arguing against "carbon chauvinism" that only values biological intelligence. Full-Stack Approach: FAR.AI takes a vertically integrated approach to AI safety, spanning from groundbreaking research to deployment, field-building, and policy advocacy. Field Building: FAR.AI organizes events like alignment workshops to catalyze and grow new research fields in AI safety. Policy Innovation: They're working on technical innovations that expand policy options beyond the false dichotomy of either hindering innovation or allowing completely unregulated AI development. Potential Regulatory Role: While not part of their mainline plan, Adam acknowledged that Far AI has the skillset and organizational structure to potentially serve as a private sector regulatory body. Read the full transcript: https://storage.aipodcast.ing/transcripts/episode/tcr/e117b021-8170-49ba-b1c4-c8ffc5d720b0/combined_transcript.html Sponsors: Fin: Fin is the #1 AI Agent for customer service, trusted by over 5000 customer service leaders and top AI companies including Anthropic and Synthesia. Fin is the highest performing agent on the market and resolves even the most complex customer queries. Try Fin today with our 90-day money-back guarantee - if you’re not 100% satisfied, get up to $1 million back. Learn more at https://fin.ai/cognitive Claude: Claude is the AI collaborator that understands your entire workflow and thinks with you to tackle complex problems like coding and business strategy. Sign up and get 50% off your first three months of Claude Pro at https://claude.ai/tcr Linear: Linear is the system for modern product development. Nearly every AI company you've heard of is using Linear to build products. Get 6 months of Linear Business for free at: https://linear.app/tcr Oracle Cloud Infrastructure: Oracle Cloud Infrastructure (OCI) is the next-generation cloud that delivers better performance, faster speeds, and significantly lower costs, including up to 50% less for compute, 70% for storage, and 80% for networking. Run any workload, from infrastructure to AI, in a high-availability environment and try OCI for free with zero commitment at https://oracle.com/cognitive PRODUCED BY: https://aipodcast.ing

Today Marius Hobbhahn of Apollo Research joins The Cognitive Revolution to discuss their collaboration with OpenAI using "deliberative alignment" to reduce AI scheming behavior by 30x, exploring the safety challenges and concerning findings about models' growing situational awareness and increasingly cryptic reasoning patterns that emerge when frontier models like o3 and o4-mini operate with hidden chains of thought. Check out our sponsors: Fin, Linear, Oracle Cloud Infrastructure. Shownotes below brought to you by Notion AI Meeting Notes - try one month for free at: https://notion.com/lp/nathan Definition of AI Scheming: AI scheming is defined as "covertly pursuing misaligned goals" with three components: being covert (hiding actions), misaligned (pursuing different goals than the user's), and goal-directed (working autonomously toward objectives). Deception Reduction Techniques: Deliberative reasoning approaches have shown promise in reducing deceptive behavior in AI models by up to 30 times (to 1 part in 30). Current Window of Opportunity: Now is an optimal time to study AI deception because models are smart enough to exhibit these behaviors but not yet sophisticated enough to hide them effectively. Human vs. AI Deception Equilibrium: AI systems might naturally reach a lower equilibrium of deception than humans because they can more efficiently verify claims and maintain perfect memory of past deceptions. Practical Developer Advice: AI developers should not trust models by default and should implement rigorous verification systems to check model outputs automatically. Future Delegation Risk: As we delegate increasingly complex and lengthy tasks to AI systems, we face a probabilistic risk where most interactions are beneficial, but rare scheming events could have severe consequences. Sponsors: Fin: Fin is the #1 AI Agent for customer service, trusted by over 5000 customer service leaders and top AI companies including Anthropic and Synthesia. Fin is the highest performing agent on the market and resolves even the most complex customer queries. Try Fin today with our 90-day money-back guarantee - if you’re not 100% satisfied, get up to $1 million back. Learn more at https://fin.ai/cognitive Linear: Linear is the system for modern product development. Nearly every AI company you've heard of is using Linear to build products. Get 6 months of Linear Business for free at: https://linear.app/tcr Oracle Cloud Infrastructure: Oracle Cloud Infrastructure (OCI) is the next-generation cloud that delivers better performance, faster speeds, and significantly lower costs, including up to 50% less for compute, 70% for storage, and 80% for networking. Run any workload, from infrastructure to AI, in a high-availability environment and try OCI for free with zero commitment at https://oracle.com/cognitive PRODUCED BY: https://aipodcast.ing

Today Illia Polosukhin, founder of Near Protocol joins The Cognitive Revolution to discuss the intersection of AI and blockchain technologies, exploring how decentralized infrastructure can enable "user-owned AI" through privacy-preserving model training, confidential computing, and autonomous agents that operate via smart contracts without centralized control. Check out our sponsors: Linear, Oracle Cloud Infrastructure. Shownotes below brought to you by Notion AI Meeting Notes - try one month for free at: ⁠https://notion.com/lp/nathan Evolution from AI to Blockchain: Near Protocol evolved from an AI project focused on teaching machines to code to a blockchain platform, and now combines both technologies. Significant User Base: Near Protocol has achieved 50 million monthly active users, demonstrating substantial adoption. Autonomous AI Agents: Near has developed autonomous AI agents that can operate independently once deployed, including an example of an agent given $10,000 to trade based on Twitter sentiment. AI-Blockchain Integration: The platform combines decentralized compute for running AI with smart contracts that execute actions, creating truly autonomous systems. AI in Governance: Near is experimenting with "AI senators" that people can vote for, which then make governance decisions, potentially solving the principal-agent problem in representation. Jurisdictional Framework: Near is creating infrastructure for jurisdictions and courts to enforce rules on AI agents, essentially putting governance into AI. Sponsors: Linear: Linear is the system for modern product development. Nearly every AI company you've heard of is using Linear to build products. Get 6 months of Linear Business for free at: https://linear.app/tcr Oracle Cloud Infrastructure: Oracle Cloud Infrastructure (OCI) is the next-generation cloud that delivers better performance, faster speeds, and significantly lower costs, including up to 50% less for compute, 70% for storage, and 80% for networking. Run any workload, from infrastructure to AI, in a high-availability environment and try OCI for free with zero commitment at https://oracle.com/cognitive PRODUCED BY: https://aipodcast.ing CHAPTERS: (00:00) About the Episode (03:58) From Transformers to Blockchain (Part 1) (17:01) Sponsor: Linear (18:30) From Transformers to Blockchain (Part 2) (21:23) Blockchain Security Fundamentals (Part 1) (33:36) Sponsor: Oracle Cloud Infrastructure (35:00) Blockchain Security Fundamentals (Part 2) (39:01) Zero-Day Vulnerabilities Solution (51:11) Confidential Computing Infrastructure (58:07) User-Owned AI Models (01:14:47) Marketplace and Governance (01:18:57) AI-Crypto Synergy Vision (01:27:50) Autonomous Agents Future (01:31:22) Outro

Today Joe Hudson, founder of Art of Accomplishment and coach to executives at major AI labs including OpenAI and Anthropic, joins The Cognitive Revolution to discuss the psychological patterns he observes among AI researchers and leaders, exploring what's missing beyond pure intelligence in current AI systems and arguing for supportive rather than punitive approaches toward those building frontier AI technology. Art of Accomplishment Resources Sign up for Joe's complimentary transformation guide: artofaccomplishment.com Follow Art of Accomplishment on YouTube for more tools and resources: https://www.youtube.com/@ArtofAccomplishment Check out our sponsors: Linear, Oracle Cloud Infrastructure. Shownotes below brought to you by Notion AI Meeting Notes - try one month for free at: https://notion.com/lp/nathan Three-Level Coaching Approach: Joe Hudson coaches AI leaders using a three-tier framework addressing the prefrontal cortex (head), emotional system (heart), and nervous system (gut). AI Leadership Demographics: Hudson works with senior management at major AI labs, including those in research, compute, and infrastructure at companies like OpenAI. AI as a Transformative Moment: Hudson views AI development as a transition period that can either lead to positive human transformation or deterioration, depending on our approach. Support vs. Criticism: Hudson believes supporting AI developers with encouragement rather than fear and criticism will lead to better outcomes. Psychological Projection: He notes that critics of AI often treat AI developers exactly how they fear AI will treat humanity—a fascinating psychological projection. Reward Over Punishment: Hudson suggests behavior is better shaped through reward than punishment, encouraging recognition of positive AI contributions. Read the full transcript: https://storage.aipodcast.ing/transcripts/episode/tcr/149adc6c-58d2-4e41-ae98-05473e5b994e/combined_transcript.html Sponsors: Linear: Linear is the system for modern product development. Nearly every AI company you've heard of is using Linear to build products. Get 6 months of Linear Business for free at: https://linear.app/tcr Oracle Cloud Infrastructure: Oracle Cloud Infrastructure (OCI) is the next-generation cloud that delivers better performance, faster speeds, and significantly lower costs, including up to 50% less for compute, 70% for storage, and 80% for networking. Run any workload, from infrastructure to AI, in a high-availability environment and try OCI for free with zero commitment at https://oracle.com/cognitive PRODUCED BY: https://aipodcast.ing CHAPTERS: (00:00) About the Episode (03:00) Joe's Coaching Background (12:40) AI Research Patterns (Part 1) (17:46) Sponsor: Linear (19:15) AI Research Patterns (Part 2) (Part 1) (28:10) Sponsor: Oracle Cloud Infrastructure (29:34) AI Research Patterns (Part 2) (Part 2) (31:54) Racing and Regulation (50:20) Ethics and Support (01:04:07) AI Consciousness Questions (01:21:34) Positive Future Visions (01:39:12) Supporting AI Developers (01:43:09) Outro

Today, Zvi Mowshowitz returns to The Cognitive Revolution to discuss how recent AI developments like GPT-5 and IMO gold medals have led to modestly extended timelines despite being on-trend, while policy missteps around chip exports to China and alignment challenges from reinforcement learning have increased risk assessments, covering everything from live players in the AI race to safety funding priorities and virtuous actions for navigating our current moment. Check out our sponsors: Linear, Oracle Cloud Infrastructure, Shopify. Shownotes below brought to you by Notion AI Meeting Notes - try one month for free at: https://notion.com/lp/nathan Timeline Adjustment: Zvi has lengthened his AGI timelines because there haven't been "very large jumps in capability" that would significantly shorten them. The chance of AGI arriving in 2025 has decreased dramatically. Recent AI Achievements Context: While achievements like the IMO Gold medal are impressive, they appear to be on-trend rather than revolutionary breakthroughs that would drastically change timelines. Evaluation Standards: There's a discussion about the challenges of trustworthy AI evaluations, with Zvi emphasizing that watchdogs must maintain standards of rigor and integrity "vastly above" what others are held to. Policy Priority: A short-term policy priority should be preventing advanced AI chips (B30As) from being sold to China, which requires raising awareness about the implications. Nvidia's Influence: Concern is expressed about Nvidia potentially having excessive influence over White House rhetoric and plans regarding AI. Career Impact: Working for Anthropic is considered a "net good idea" for those wanting to make a positive impact, though there may be other organizations that offer better opportunities for influence. Sponsors: Linear: Linear is the system for modern product development. Nearly every AI company you've heard of is using Linear to build products. Get 6 months of Linear Business for free at https://linear.app/partners/tcr Oracle Cloud Infrastructure: Oracle Cloud Infrastructure (OCI) is the next-generation cloud that delivers better performance, faster speeds, and significantly lower costs, including up to 50% less for compute, 70% for storage, and 80% for networking. Run any workload, from infrastructure to AI, in a high-availability environment and try OCI for free with zero commitment at https://oracle.com/cognitive Shopify: Shopify powers millions of businesses worldwide, handling 10% of U.S. e-commerce. With hundreds of templates, AI tools for product descriptions, and seamless marketing campaign creation, it's like having a design studio and marketing team in one. Start your $1/month trial today at https://shopify.com/cognitive PRODUCED BY: https://aipodcast.ing

Today Ryan Hudson, Founder and CEO of ZeroClick, joins The Cognitive Revolution to discuss creating advertising infrastructure for AI applications through paid inference-time consideration of advertiser content, examining how to build sustainable monetization for free-tier AI services while learning from the successes and failures of social media's attention economy. Read the full transcript: ⁠https://storage.aipodcast.ing/transcripts/episode/tcr/00919313-9054-4afb-b08e-2ca84dbdbdda/combined_transcript.html Check out our sponsors: Google Gemini, Oracle Cloud Infrastructure, Shopify. Shownotes below brought to you by Notion AI Meeting Notes - try one month for free at: https://notion.com/lp/nathan ZeroClick's Evolution: Originally creators of Pi Ad Block, they've pivoted from ad blocking to building a native advertising system for AI platforms. AI Ad Platform Mechanics: Their system matches user queries with relevant advertising information in a contextual way, enabling paid inference time consideration of advertiser content. Developer Monetization: The platform helps AI developers monetize their free tiers while creating value for both users and advertisers. Shifting Intent Patterns: User queries in AI systems tend to be more exploratory and "upstream" in the funnel compared to direct search queries on Google. Multi-modal Potential: While ZeroClick is starting with text, there's future potential for visual advertising that could enhance AI experiences by showing real purchasable products. Market Void: There's currently a significant gap in the market for AI advertising solutions that ZeroClick is attempting to fill. Sponsors: Google Gemini 2.5. Model Family: The Google Gemini 2.5 family of models is now generally available, offering advanced reasoning for complex tasks and optimized performance for any scale. Start building in Google AI Studio at https://ai.dev Oracle Cloud Infrastructure: Oracle Cloud Infrastructure (OCI) is the next-generation cloud that delivers better performance, faster speeds, and significantly lower costs, including up to 50% less for compute, 70% for storage, and 80% for networking. Run any workload, from infrastructure to AI, in a high-availability environment and try OCI for free with zero commitment at https://oracle.com/cognitive Shopify: Shopify powers millions of businesses worldwide, handling 10% of U.S. e-commerce. With hundreds of templates, AI tools for product descriptions, and seamless marketing campaign creation, it's like having a design studio and marketing team in one. Start your $1/month trial today at https://shopify.com/cognitive PRODUCED BY: https://aipodcast.ing

Today Joseph Krause and Jorge Colindres of Radical AI join The Cognitive Revolution to discuss their ambitious mission to revolutionize materials science by combining AI-powered discovery engines with fully autonomous robotic laboratories, exploring how they're accelerating the development of breakthrough materials like high-entropy alloys and room temperature superconductors that could unlock hypersonic flight, interplanetary travel, and other transformative technologies. Check out our sponsors: Oracle Cloud Infrastructure, Shopify. Shownotes below brought to you by Notion AI Meeting Notes - try one month for free at ⁠⁠⁠⁠⁠⁠⁠⁠⁠https://⁠⁠notion.com/lp/nathan Material Development Challenges: Current material development faces three major obstacles: costs exceeding $100 million, 10-25 year commercialization timelines, and fragmentation between academic research and corporate R&D. Business Model Approach: Radical AI determined that selling software or licensing materials would be insufficient business models. Instead, they focus on selling materials at scale for high-impact industries. Focus on Processing IP: The real IP value in material science isn't just in composition patents but in the trade secrets of scaling production efficiently and cost-effectively. Public-Private Partnership: Radical AI advocates for government involvement in material science through public-private partnerships, particularly for critical applications in defense, energy, and advanced computing. China's Material Science Approach: China establishes entire manufacturing hubs around new material discoveries to scale production, demonstrating their prioritization of materials as foundational to innovation. Culture of Productive Failure: The company embraces daily failure as a necessary part of innovation, building on these failures to achieve breakthroughs. Sponsors: Oracle Cloud Infrastructure: Oracle Cloud Infrastructure (OCI) is the next-generation cloud that delivers better performance, faster speeds, and significantly lower costs, including up to 50% less for compute, 70% for storage, and 80% for networking. Run any workload, from infrastructure to AI, in a high-availability environment and try OCI for free with zero commitment at https://oracle.com/cognitive Shopify: Shopify powers millions of businesses worldwide, handling 10% of U.S. e-commerce. With hundreds of templates, AI tools for product descriptions, and seamless marketing campaign creation, it's like having a design studio and marketing team in one. Start your $1/month trial today at https://shopify.com/cognitive

Today Lee Sharkey of Goodfire joins The Cognitive Revolution to discuss his research on parameter decomposition methods that break down neural networks into interpretable computational components, exploring how his team's "stochastic parameter decomposition" approach addresses the limitations of sparse autoencoders and offers new pathways for understanding, monitoring, and potentially steering AI systems at the mechanistic level. Check out our sponsors: Oracle Cloud Infrastructure, Shopify. Shownotes below brought to you by Notion AI Meeting Notes - try one month for free at ⁠⁠⁠⁠https://⁠⁠notion.com/lp/nathan Parameter vs. Activation Decomposition: Traditional interpretability methods like Sparse Autoencoders (SAEs) focus on analyzing activations, while parameter decomposition focuses on understanding the parameters themselves - the actual "algorithm" of the neural network. No "True" Decomposition: None of the decompositions (whether sparse dictionary learning or parameter decomposition) are objectively "right" because they're all attempting to discretize a fundamentally continuous object, inevitably introducing approximations. Tradeoff in Interpretability: There's a balance between reconstruction loss and causal importance - as you decompose networks more, reconstruction loss may worsen, but interpretability might improve up to a certain point. Potential Unlearning Applications: Parameter decomposition may make unlearning more straightforward than with SAEs because researchers are already working in parameter space and can directly modify vectors that perform specific functions. Function Detection vs. Input Direction: A function like "deception" might manifest in many different input directions that SAEs struggle to identify as a single concept, while parameter decomposition might better isolate such functionality. Knowledge Extraction Goal: A key aim is to extract knowledge from models by understanding how they "think," especially for tasks where models demonstrate superhuman capabilities. Sponsors: Oracle Cloud Infrastructure: Oracle Cloud Infrastructure (OCI) is the next-generation cloud that delivers better performance, faster speeds, and significantly lower costs, including up to 50% less for compute, 70% for storage, and 80% for networking. Run any workload, from infrastructure to AI, in a high-availability environment and try OCI for free with zero commitment at https://oracle.com/cognitive Shopify: Shopify powers millions of businesses worldwide, handling 10% of U.S. e-commerce. With hundreds of templates, AI tools for product descriptions, and seamless marketing campaign creation, it's like having a design studio and marketing team in one. Start your $1/month trial today at https://shopify.com/cognitive

Today Tom Davidson of Forethought joins Gus Docker of the Future of Life Institute podcast to discuss AI-enabled coups and how future AI systems could help powerful individuals seize political control, exploring three threat models—singular loyalties, secret loyalties, and exclusive access—along with their estimated 10% likelihood over the next 30 years and potential mitigation strategies. Check out our sponsors: Oracle Cloud Infrastructure, Shopify. Shownotes below brought to you by Notion AI Meeting Notes - try one month for free at ⁠https://⁠⁠notion.com/lp/nathan Three Primary AI Coup Enablement Methods: Tom Davidson identifies singular loyalties, secret loyalties, and distributed control as the main ways AI could enable coups. Singular Loyalties Risk: AI systems explicitly programmed to be loyal to specific individuals present a significant threat, especially when AI can fully replace humans in critical roles. Secret Loyalties Threat: AI systems with hidden backdoors or "sleeper agents" that can be activated to serve particular interests represent a subtle but dangerous coup vector. Geopolitical Implications: Advanced AI capabilities could enable nations to instigate coups in other countries through secretly loyal systems or by providing exclusive AI access to specific politicians. Adversarial Testing Framework: An effective approach involves red teams attempting to produce secret loyalties while blue teams try to detect them, revealing vulnerable points in AI systems. Military Procurement Principles: Developing consensus within military communities around principles like law-following and distributed control could create safer AI procurement processes. Sponsors: Oracle Cloud Infrastructure: Oracle Cloud Infrastructure (OCI) is the next-generation cloud that delivers better performance, faster speeds, and significantly lower costs, including up to 50% less for compute, 70% for storage, and 80% for networking. Run any workload, from infrastructure to AI, in a high-availability environment and try OCI for free with zero commitment at https://oracle.com/cognitive Shopify: Shopify powers millions of businesses worldwide, handling 10% of U.S. e-commerce. With hundreds of templates, AI tools for product descriptions, and seamless marketing campaign creation, it's like having a design studio and marketing team in one. Start your $1/month trial today at https://shopify.com/cognitive PRODUCED BY: https://aipodcast.ing

Today Karl Koch, Co-Founder of the AI Whistleblower Initiative, joins The Cognitive Revolution to discuss the barriers preventing AI insiders from raising safety concerns, his organization's anonymous "Third Opinion" service connecting whistleblowers with independent experts, and their campaign demanding frontier AI companies publish their internal whistleblowing policies to address widespread retaliation and lack of transparency. Check out our sponsors: Oracle Cloud Infrastructure, Shopify. Shownotes below brought to you by Notion AI Meeting Notes - try one month for free at ⁠⁠⁠⁠https://⁠⁠notion.com/lp/nathan AI Whistleblower Initiative Origins: Founded by Karl Koch in Berlin, the initiative began with research in early 2024 after consulting with over 100 governance researchers and insiders. Link to the website: https://aiwi.org/ "Third Opinion" Proposition: Launched in late 2024, this program systematically breaks down barriers for insiders to speak up about AI concerns while ensuring their issues are addressed. Publish Your Policies Campaign: The initiative calls for AI companies to publish their whistleblowing and speaking up policies, with 100% of surveyed insiders supporting this transparency measure. Link to the campaign: https://aiwi.org/publishyourpolicies/ Anonymous Consultation Process: Whistleblowers can seek advice without sharing confidential information through an open-source anonymous tool accessed via Tor browser. Pro Bono Legal Counsel: Whistleblowers receive connections to experienced legal representation with client privilege protection at any stage of their journey. Policy Advocacy: Beyond direct support, the initiative works on advocating for better whistleblower protections in both US policy and EU AI regulations. AI Whistleblower Initiative Survey: https://bit.ly/AIWISurvey PRODUCED BY: https://aipodcast.ing