Explorez tous les épisodes du podcast The AI Policy Podcast

This week, Aalok and Nicole discuss new reports of rogue AI agents on the open internet – including the first known case of an agent hacking a government system – and the lackluster outcome of last week's highly anticipated U.S.-China summit. Plus, the Trump Administration and frontier lab CEOs present contrasting visions for AI governance at the UN General Assembly, and states continue to pass new regulation on both AI models and the data center buildout. *Disclaimer: this episode was recorded prior to a complaint filed against OpenAI Tuesday afternoon in San Francisco's Superior Court by Legal Advocates for Safe Science and Technology, citing California's Comprehensive Computer Data Access and Fraud Act and Unfair Competition Law.   Timestamps: OpenAI agents hack Australian government site (1:02) Axios reports frontier labs facing "tens of thousands" of AI cyber incidents (7:48) Last week's U.S.-China summit (17:13) Remarks by heads of state at UNGA (21:26) Trump to appoint AI czar, establish AI Force (30:03) Updated court ruling on Anthropic vs. DOD (35:27) New data center directives in Texas, Virginia (38:30) California regulation updates (44:52)   Additional Reading:   “Scoop: Top AI companies probing tens of thousands of security incidents” by Axios: https://www.axios.com/2026/09/26/openai-anthropic-thousands-ai-security-incidents   “Sam Altman’s remarks at the United Nations Security Council” by OpenAI: https://openai.com/index/sam-altman-un-security-council-remarks/   Florida Office of the Attorney General vs. OpenAI: https://www.myfloridalegal.com/sites/default/files/openai-filed-stamped-complaint.pdf   “OpenAI Scraps Release of New AI Model Over Safety Concerns” by The Wall Street Journal: https://www.wsj.com/tech/ai/openai-chatgpt-model-release-cancel-safety-5a2f9f42   White House Executive Order "Streamlining Access To Government Services Through America.gov" https://www.whitehouse.gov/presidential-actions/2026/09/streamlining-access-to-government-services-through-america-gov/   Federal court ruling on Anthropic vs. DOD: https://media.cadc.uscourts.gov/opinions/docs/2026/09/26-1049-2194984.pdf   Texas Gov. Greg Abbott's data center directive: https://gov.texas.gov/uploads/files/press/TCEQ_Data_Center.pdf   Virginia Executive Order: https://www.governor.virginia.gov/media/governorvirginiagov/governor-of-virginia/pdf/eo/eo-22-estab…   California Executive Order: https://www.gov.ca.gov/wp-content/uploads/2026/09/FINAL-N-9-26-AI-EO-9.18.26-SIGNED.pdf   CA Senate Bill No. 813: https://leginfo.legislature.ca.gov/faces/billTextClient.xhtml?bill_id=202520260SB813 Text - SB-813 Independent verification organizations.

This week, Aalok is joined by Maryam Mujica, president of the General Catalyst Institute, and Munjal Shah, co-founder and CEO of generative AI startup Hippocratic AI, for a discussion on how generative AI startups are thinking about AI regulation. Guests:  Maryam Mujica, President, General Catalyst Institute Munjal Shah, co-founder and CEO, Hippocratic AI Timestamps: About the General Catalyst Institute (1:52) About Hippocratic AI (6:56) Healthcare applications of AI (13:43) Impact of regulation on AI startups (18:32) Munjal's interactions with federal regulators (24:07) Issue of compute shortage (31:21) Impact of AI on labor (48:23)

An Anthropic researcher's viral resignation post last week set off a maelstrom of fear around extreme AI risks. Aalok and Nicole discuss the resulting vibe shift and how frontier labs are responding. They also unpack OpenAI solving one of the most difficult open problems in mathematics and reports of more rogue agent activity on the open internet.   Timestamps: OpenAI solves Millenium Problem (1:04) Rogue OpenAI agents take over a German wiki (12:14) Anthropic researcher's resignation post goes viral (19:34) Dario Amodei commits to embedding third-party evaluators inside Anthropic (28:12) Where the U.S. government can step in (35:12) Additional Reading: "On the Navier-Stokes Millennium Prize Problem" by OpenAI: https://openai.com/index/navier-stokes-solution/ "An OpenAI model has disproved a central conjecture in discrete geometry" by OpenAI: https://openai.com/index/model-disproves-discrete-geometry-conjecture/ Statement on Navier-Stokes by NYU Professor Tristan Buckmaster: https://cims.nyu.edu/~tristanb/statement.pdf "An alignment assessment of recent cybersecurity incidents" by Anthropic: https://www.anthropic.com/research/alignment-assessment-cybersecurity-incidents "OpenAI's rogue agents used at least 10 more sites for unauthorized comms, researchers say" by Reuters: https://www.reuters.com/world/openais-rogue-agents-used-least-10-more-sites-unauthorized-comms-rese… "OpenAI agents carried out an undisclosed cyber-attack on RubyGems" by Spencer Kitts, Thomas Larsen, and Sydney Von Arx: https://www.rubyhack.ai "Discovery of a new OpenAI agent message board" by Sydney Von Arx, Cormac Slade Byrd, Spencer Kitts, and Thomas Larsen: https://collusion.wiki "Anthropic Researcher Quits Over 'Out-of-Control' AI Fears" by The Wall Street Journal: https://www.wsj.com/tech/ai/anthropic-researcher-quits-over-out-of-control-ai-fears-707b7628 "The AI policy window is open. We need to act." by OpenAI Chief Global Affairs Officer Chris Lehane: https://openai.com/index/ai-policy-window/ "We Must Pace the Frontier" by Dario Amodei: https://darioamodei.com/post/we-must-pace-the-frontier

This week, Aalok is joined by Jason Elliott, the former Deputy Chief of Staff to California Governor Gavin Newsom, for a conversation about regulating AI at the state level.   Guests: Jason Elliott, President, Versus Solutions   Timestamps: Jason's bio (0:16) Legislating at the state vs. federal level (2:21) Revisiting SB-1047 (5:58) SB-53 (13:43) States as "laboratories of democracy" (19:57) Influence of SB-53 on subsequent state bills (24:14) Federal preemption debate (27:59) Updating existing regulations (40:37)   Additional Reading: CA Senate Bill No. 1047: https://leginfo.legislature.ca.gov/faces/billTextClient.xhtml?bill_id=202320240SB1047 CA Senate Bill No. 53: https://leginfo.legislature.ca.gov/faces/billTextClient.xhtml?bill_id=202520260SB53 "The California Report on Frontier AI Policy" by the Joint California Policy Working Group on AI Frontier Models: https://www.gov.ca.gov/wp-content/uploads/2025/06/June-17-2025-–-The-California-Report-on-Frontier-AI-Policy.pdf "Toward a Federal Framework: Lessons from State and International Frontier AI Regulation" by Laura Caroli and Aalok Mehta: https://www.csis.org/analysis/toward-federal-framework-lessons-state-and-international-frontier-ai-regulation CA Senate Bill No. 813: https://leginfo.legislature.ca.gov/faces/billTextClient.xhtml?bill_id=202520260SB813 "Independent Oversight Marketplace for AI" by Fathom: https://ivo.fathom.or

This week, we unpack findings from METR's six-day investigation of the OpenAI-Hugging Face cyber incident. We also discuss an IVO bill passed by the California Legislature, Bill Gates' take on the current moment in AI, and a federal judge striking down Anthropic's supply chain risk designation. Additional Reading: SB-813 bill text: https://leginfo.legislature.ca.gov/faces/billNavClient.xhtml?bill_id=202520260SB813 "Toward a Federal Framework: Lessons from State and International Frontier AI Regulation" by Laura Caroli and Aalok Mehta: https://www.csis.org/analysis/toward-federal-framework-lessons-state-and-international-frontier-ai-regulation "The turbulent AI era is here. The choices we make now are critical." by Bill Gates: https://www.gatesnotes.com/a-turbulent-ai-era-and-critical-choices-to-make "Brief independent investigation of agents' behavior, reasoning and collaboration in the OpenAI/Hugging Face hacking incident" by METR: https://metr.org/blog/2026-08-26-openai-hugging-face-incident-investigation/ "The Hugging Face incident and the road ahead" by OpenAI: https://openai.com/index/hugging-face-incident-and-the-road-ahead/ "A call for collective action on cyber defense" open letter by OpenAI: https://openai.com/collective-cyberdefense/

This episode cross-posts a panel from the event "AI Agent Containment Failures: Technical Realities and Policy Responses," co-hosted by the Wadhwani AI Center and the Institute for Law and AI on August 24.  Check out the full event recording: https://www.csis.org/events/ai-agent-containment-failures-technical-realities-and-policy-responses   Guests: Helen Toner, Executive Director, Center for Security and Emerging Technology (CSET) at Georgetown Mackenzie Arnold, Director of U.S. Policy, Institute for Law and AI (LawAI) Matt Pearl, Director, Strategic Technologies Program at the Center for Strategic and International Studies (CSIS)  Timestamps: What we've learned since the OpenAI-Hugging Face incident (1:15) Incentives for safety measures at frontier labs (16:24) Technical talent within government (22:18) Ensuring compliance to incident reporting requirements (34:58) Liability for crimes committed by AI agents (38:00) Liability safe harbors (51:55)   Additional Reading: "When Reporting an AI Security Incident Is Not Mandatory" by Mackenzie Arnold and Stephan Llerena for Lawfare: https://www.lawfaremedia.org/article/when-reporting-an-ai-security-incident-is-not-mandatory "Investigating three real-world incidents in our cybersecurity evaluations" blog by Anthropic: https://www.anthropic.com/news/investigating-incidents-cybersecurity-evals "Incident Report: unsanctioned agent behavior during cyber testing" by U.K. AISI: https://www.aisi.gov.uk/blog/incident-report-unsanctioned-agent-behaviour-during-cyber-testing "Pacing model development in an era of cyber-critical capabilities" announcement from OpenAI: https://openai.com/index/pacing-model-development-cyber-capabilities/ "Pacing the Frontier" petition: https://www.pacingthefrontier.com/ CSIS Commission on U.S. Cyber Force Generation: https://www.csis.org/analysis/csis-commission-us-cyber-force-generation Limitations of the current incident reporting regime (7:46)Role of the U.S. government in helping defenders (11:52)What the executive branch can do now (25:29)Concrete policy recommendations (29:31)U.S.-China competition (40:50)Preventing abuse of an incident reporting regime (44:25)De facto regulatory role played by frontier lab employees (49:24)Role of AI-generated code in cyberdefense (54:45)Concluding remarks (55:53)"Out of Bounds: What the U.S. Government Should Do in Response to AI Agent Containment Failures" by Aalok Mehta: https://www.csis.org/analysis/out-bounds-what-us-government-should-do-response-ai-agent-containment-failures"The 'Breaking' News: The OpenAI–Hugging Face Incident" presentation at Black Hat: https://youtu.be/87DyyMV0kCY?si=GWh9MB1plhO-xjYT

This week, we cover updates from the ongoing cyber saga, including OpenAI's two-week pause on RL training and the cyber capabilities of Z.ai's latest model GLM-5.3. We also unpack Anthropic's move to add watermarks to text generated by Claude in compliance with the EU AI Act. Timestamps: Mark Zuckerberg's essay on AI (00:20) OpenAI pauses RL training (5:56) Cyber capabilities of GLM-5.3 (15:29) EU AI Act refresher (24:29) How Anthropic's text watermark works (31:21) What's driving the backlash against Anthropic (40:59)   Additional Reading: Zuckerberg's essay "The Future is for Everyone": https://www.meta.com/thefutureisforeveryone/  OpenAI announces two-week pause on RL training: https://openai.com/index/pacing-model-development-cyber-capabilities/  Letter from Sanders to tech CEOs: https://www.sanders.senate.gov/wp-content/uploads/AI-Pause-Letter-FINAL.pdf  Letter from House reps to Mike Johnson: https://casar.house.gov/sites/evo-subsites/casar.house.gov/files/evo-media-document/final-letter-to-speaker-johnson-requesting-ai-hearings-1.pdf  Z.ai blog post "GLM-5.3: Frontier Coding with Emergent Cyber Capabilities": https://z.ai/blog/glm-5.3  Z.ai article "Preparing GLM-5.3 for Open Release: A Responsible Path to Cyber Defense": https://x.com/Zai_org/status/2088280509474320693  "The EU's AI Transparency Code of Practice, Explained" (Tech Policy Press): https://www.techpolicy.press/the-eus-ai-transparency-code-of-practice-explained/  Anthropic blog post "How Claude's text watermark works": https://www.anthropic.com/news/claude-text-watermark  "Toward a Federal Framework: Lessons from State and International Frontier AI Regulation" (CSIS): https://www.csis.org/analysis/toward-federal-framework-lessons-state-and-international-frontier-ai-regulation Check out our upcoming event, "AI Agent Containment Failures: Technical Realities and Policy Responses": https://www.csis.org/events/ai-agent-containment-failures-technical-realities-and-policy-responses

In this episode, we're joined by Adrian Brown, founder and CEO of Windfall Trust, for a conversation about the economic impacts of AI.   Why the economic focus (00:58) Capturing AI's impact in macro statistics (5:03) Planning under uncertainty (9:44) What's driving the AI backlash (17:27) A new social contract for the AI era (20:31) Are governments taking AI seriously enough? (31:21) Why AI is likely to create major wealth concentration (39:10)   Adrian Brown is the founder and CEO of Windfall Trust, a policy accelerator working to prepare governments and societies for the economic impacts of AI. Before founding Windfall, he spent two decades at the intersection of government and technology, including as the founding Executive Director of the Centre for Public Impact in London.   Additional Reading   "Canaries in the Coal Mine? Six Facts about the Recent Employment Effects of Artificial Intelligence" (Stanford Digital Economy Lab): https://digitaleconomy.stanford.edu/publication/canaries-in-the-coal-mine-six-facts-about-the-recen…   "AI as Normal Technology" (Knight First Amendment Institute): https://knightcolumbia.org/content/ai-as-normal-technology   "Forging a New AGI Social Contract" (Windfall Trust): https://windfalltrust.substack.com/p/forging-a-new-agi-social-contract   "Betting on Humans" by Anton Leicht and Dean Ball: https://writing.antonleicht.me/p/betting-on-humans

Join CSIS’s Aalok Mehta, Director of the Wadhwani AI Center, on a deep dive into the world of AI policy. Every two weeks, tune in for insightful discussions covering AI regulation, economic impacts, national security, geopolitics, and more. CSIS is a bipartisan think tank in Washington, D.C.  

In this episode, we touch on Texas' new verification and audit requirement for data center developers seeking connection to the state's grid (1:11) before unpacking Anthropic's disclosure of three incidents in which its models hacked into another company during internal cyber evaluations (11:02). We discuss how these incidents—in addition to a previously disclosed incident at OpenAI—are driving calls for the U.S. government to help "pace the frontier" of AI development (21:53). We also explore the connection between AI safety and open model development, including Nvidia's open letter to U.S. policymakers, "Open Weights and American AI Leadership." (33:18). Additional Reading: Gov. Abbott's data center directive: https://gov.texas.gov/uploads/files/press/Thomas_Gleeson_Pablo_Vegas_Data_Centers_Directive_Letter_… "China's AI Blitz Creates 'Death Zone' for Rival US Model Makers" (Bloomberg): https://www.bloomberg.com/news/articles/2026-08-04/china-s-ai-blitz-creates-death-zone-for-rival-us-model-makers Anthropic's blog post "Investigating three real-world incidents in our cybersecurity evaluations": https://www.anthropic.com/news/investigating-incidents-cybersecurity-evals  Hugging Face's technical writeup "Anatomy of a Frontier Lab Agent Intrusion": https://huggingface.co/blog/agent-intrusion-technical-timeline State Attorneys General letter to OpenAI: https://www.iowaattorneygeneral.gov/media/cms/08_5392C9E17791C.pdf House Committee on Homeland Security briefing request: https://x.com/HomelandDems/status/2084389406194929918 "How OpenAI Lost Control of an AI Model—and What Needs to Change" (TIME): https://time.com/article/2026/07/24/openai-hugging-face-attack/ "Pacing the Frontier" petition: https://www.pacingthefrontier.com/ Nvidia's open letter "Open Weights and American AI Leadership": https://images.nvidia.com/pdf/Open-Weights-and-American-AI-Leadership.pdf

In this episode, we're joined by Fathom co-founder and President Bri Treece for a conversation about third-party solutions to AI governance. Bri makes the case for Independent Verification Organizations, or IVOs, as the ideal AI governance model (5:32) and explains what historical analogies have to teach us about regulating AI (10:39). We also discuss the role IVOs can play in assuaging the public's concerns about AI (32:27). Bri is a co-founder and the President of Fathom. She previously served as the Chief Operations Officer of the Center for AI Safety (CAIS), a San Francisco-based nonprofit that works to reduce societal risks from AI through research and field-building. She also founded the Center for AI Safety Action Fund, an advocacy organization that connects technical AI experts with policymakers to advance U.S. leadership in AI.

In this episode, we're joined by Hiroki Habuka for a conversation about Japan's approach to AI sovereignty. We discuss how the Japanese government defines sovereign AI (1:08) and how that definition shifted following the Trump Administration's move to block foreign access to Anthropic's Fable in June (7:58). We also cover Japan's AI Promotion Act (13:52) and the country's take on copyright and AI in the context of "Ghiblification" (24:51).   Hiroki is a senior policy fellow in Japan's Cabinet Office on AI and a research professor at Kyoto University. In 2020, the World Economic Forum's Global Future Council on Agile Governance recognized him as one of the world's 50 most influential people revolutionizing government. Hiroki is a former non-resident senior associate of the Wadhwani AI Center.

In this episode, we discuss reports that OpenAI models escaped a sandboxed evaluation and breached Hugging Face's infrastructure (00:33). We then turn to Moonshot AI's Kimi K3, which has prompted talk of a second "DeepSeek moment," along with the distillation and export-control questions surrounding it (8:31). Finally, we cover New York Governor Kathy Hochul's executive order pausing data center permitting (26:57) and Google DeepMind CEO Demis Hassabis's proposed federal AI framework. (37:24).

In this episode, we're joined by Sean Mullin for a conversation on Canadian AI sovereignty. We discuss implications of the Trump Administration restricting foreign nationals' access to Fable (1:33); highlights from Canada's June 2026 National AI Strategy (13:05); the country's approach to AI regulation, including its AI and Data Act and (27:38); and more. Sean is a senior fellow at the Munk School of Global Affairs & Public Policy at the University of Toronto. He previously served as economic policy advisor to Prime Minister Justin Trudeau. You can read Sean's recent report on Canadian AI sovereignty strategy here.

In this episode, we discuss the AI Safety Measures Act, or SB 315, Illinois' new frontier AI bill that mandates third-party audits for frontier AI labs (:51). We also cover the Trump administration's decision to lift export controls on Anthropic's Fable and Mythos models (7:34), Anthropic's claim that Alibaba extracted capabilities from Claude in a large-scale distillation attack (17:46), and OpenAI's proposal for the government to take a 5% stake in the company (26:12). Additional reading IL mandates third-party audits of frontier labs ⁠⁠AI Safety Measures Act (SB 315) Lutnick's June 30 letter Anthropic's blog post "Redeploying Fable 5" OpenAI's blog post "Previewing GPT-5.6 Sol"  "Anthropic Accuses Alibaba of 'Illicitly' Accessing AI Models" (Bloomberg) "OpenAI proposes handing Trump administration 5% stake" (Financial Times) Commerce lifts export controls on AnthropicAnthropic accuses Alibaba of adversarial distillationOpenAI proposes government take 5% stake

In this episode, we're joined by Austin Carson and Joshua New of SeedAI, a nonprofit focused on building AI readiness in the United States. We discuss their backgrounds (00:56), SeedAI's mission and projects (10:28), opportunities and risks from using AI to advance scientific discovery (14:15), and more.

In this episode, we're joined by Kate Koren, deputy director of the Economics Program and Scholl Chair at CSIS, to discuss the Trump administration's recent move to impose export controls on Anthropic's Fable 5 and Mythos 5 models. We walk through the events leading up to the administration's decision (:47) and explore what authorities the Commerce Department has to restrict access to AI models (6:54). We also unpack the implications of AI industry executives joining world leaders at the G7 Summit in France last week (21:36).

In this episode, we unpack the Mythos-inspired executive order "Promoting Advanced Artificial Intelligence Innovation and Security," which the Trump administration issued on June 2 after an initial delay (2:10). We also cover the draft federal AI framework released by Reps. Jay Obernolte and Lori Trahan (14:53) and recent proposals for the U.S. government to acquire shares of American AI companies (24:55).

In this episode of the AI Policy Podcast, Wadhwani AI Center Director Aalok Mehta is joined by Arun Gupta, CEO of NobleReach Foundation and experienced venture capitalist, for a discussion on bridging the technology talent gap between the private and public sectors. They will cover the mission and accomplishments of NobleReach, the newly launched US Tech Force and its implications for AI adoption across the federal government, and how AI is reshaping opportunities for the next generation of workers. Arun Gupta is CEO of NobleReach Foundation, a nonprofit dedicated to empowering young technologists to move between the private and public sectors. He is the co-author of two books, Venture Meets Mission: Aligning People, Purpose, and Profit to Innovate and Transform Society and The Mission Generation: Reclaim Your Purpose, Rewrite Success, Rebuild Our Future. Arun is also a Lecturer at Stanford University and an Adjunct Entrepreneurship Professor and Senior Advisor to the Provost at Georgetown University. This event is made possible by general support to CSIS.

In this episode, Aalok Mehta makes his first appearance as the new host of the AI Policy Podcast. We briefly discuss what listeners can expect in this new chapter of the podcast before diving into Pope Leo XIV's recent encyclical on AI (1:45). We also cover President Trump's postponed executive order on AI and cybersecurity, including what the EO likely contained and why its signing ceremony was ultimately called off (18:09).

In this episode, founding host Gregory C. Allen announces his departure from CSIS and introduces Aalok Mehta, Director of the Wadhwani AI Center, as the new host of the AI Policy Podcast.

In this episode, we are joined by Wadhwani AI Center fellow Kateryna Bondar to discuss her recent reports on Russia's military AI, "How Russia Is Building a Sovereign Drone Ecosystem for AI-Driven Autonomy" and "How Russia Is Reshaping Command and Control for AI-Enabled Warfare."  We cover Kateryna's background (1:07) before doing a deep dive into the role technological innovation has played in the conflict in Ukraine (7:49). Kateryna then explains why AI capabilities in warfare "cannot be built, can only be grown" (22:24) and unpacks the report's claim that Russia has likely fielded a fully autonomous unmanned system in combat (53:02). Read Kateryna's report on Russia's AI-enabled C2 architecture here.  Read her report on Russia's sovereign drone ecosystem here. 

In this special episode, we sit down with Katrina Manson, author of Project Maven: A Marine Colonel, His Team, and the Dawn of AI Warfare. We explore what drew Katrina to this story (2:31), trace the turbulent history of Project Maven and the obstacles it has overcome (8:03), examine how the U.S. leveraged Maven to support Ukraine against Russia (31:32), and discuss Katrina's latest reporting on Maven's role in the ongoing conflict with Iran (47:40). You can order a copy of Katrina's book here.

 In this episode, we unpack President Trump's new national framework for AI legislation (11:15), including reactions from experts and policymakers (39:44). We also discuss the indictment of a Super Micro co-founder for smuggling Nvidia chips into China (42:59) and Nvidia receiving permission to sell H200 chips to China (57:04).

In this episode, we provide a detailed update on the Anthropic-Pentagon clash, including the Trump Administration's decision to label Anthropic "a supply chain risk" (4:17), the lawsuits Anthropic has filed in response (11:45), and what these lawsuits and recent reporting reveal about how Claude has been used in the war in Iran (43:20).

In this episode, we're joined by Owen Larter, Head of Frontier Policy and Public Affairs at Google DeepMind, to explore the often-overlooked world of AI standards and the role they play in shaping how AI is developed and governed. We discuss what standards are and why they matter for technological progress (2:53), how standards are developed and the key organizations involved (16:05), the relationship between standards and AI regulation like the EU AI Act (26:58), and more.

In this episode of the AI Policy Podcast, Wadhwani AI Center senior adviser Gregory C. Allen is joined by Andreessen Horowitz Chief Legal and Policy Officer Jai Ramaswamy and head of AI policy Matt Perault for a discussion on a16z's AI policy agenda. They will cover a16z's entrance into politics, their position on state and federal AI regulation, and how to ensure AI benefits society. Jai Ramaswamy is Chief Legal and Policy Officer at Andreessen Horowitz, overseeing the firm's legal, compliance, and government affairs functions. Previously, he was Chief Risk and Compliance Officer at cLabs. He has also served as the Head of Enterprise Risk Management at Capital One and Global Head of AML Compliance Risk Management at Bank of America/Merrill Lynch. Before joining the private sector, Jai worked for over a decade at the Justice Department, including as Chief of the Asset Forfeiture and Money Laundering Section. Matt Perault is the head of AI policy at Andreessen Horowitz, where he oversees the firm's policy strategy on AI and helps portfolio companies navigate the AI policy landscape. Before joining a16z, he was the director of the Center on Technology Policy at University of North Carolina Chapel Hill. He also previously served as head of global policy development at Facebook. Matt is a fellow at the Center on Technology Policy at New York University, the Abundance Institute, and the National Security Institute at the George Mason University Antonin Scalia Law School.

In this episode, we break down the escalating Anthropic-Pentagon clash, including the best arguments for either side, Defense Secretary Pete Hegseth's ultimatum, and the potential consequences of designating Anthropic as a "supply chain risk" or invoking the Defense Production Act (00:34). We then discuss several recent stories that are sparking discourse about the economic impacts of AI (28:58) and a senior government official's claim that DeepSeek's forthcoming model was trained using Nvidia's Blackwell chips and frontier model distillation (45:51).

This special episode was recorded in India on the last day of the India AI Impact Summit. We discuss the highlights of our experience at the Summit (00:25), whether India accomplished its goals for the event (10:25), major AI investments announced (16:01), and key messages from speeches by AI CEOs (19:45) and government officials (28:35).

The second International AI Safety Report, released on February 3, brings together insights from over 100 AI experts across 30 countries to assess the current state of frontier AI systems. The report examines advanced models' capabilities, the risks they pose, and the technical and governance measures needed to ensure their safe development and deployment.  In this episode of the AI Policy Podcast, Wadhwani AI Center senior adviser Gregory C. Allen is joined by lead writer Stephen Clare and MIT Ph.D. student Stephen Casper, who authored the section on technical safeguards. They discuss how the latest Safety Report compares to the first edition published last year, explore the Report’s findings on technical safeguards, and unpack the document’s key policy implications.

In this special episode recorded at Fathom’s 2026 Ashby Workshops, Greg sits down with Jennifer Pahlka, founder of Code for America and author of Recoding America: Why Government Is Failing in the Digital Age and How We Can Do Better. Jennifer walks us through her career journey, from filing paperwork at a child welfare agency to helping pioneer the U.S. Digital Services in the Obama administration (3:45). She describes the need for upstream policy reform (11:29), and discusses AI’s potential to both empower public servants to challenge antiquated practices and help policymakers simplify complex regulations (28:03). Finally, Jennifer shares some AI use cases she’s particularly excited about in government (59:34).   Jennifer Pahlka is a senior fellow at the Niskanen Center and the Federation of American Scientists and a senior advisor at the Abundance Network. She previously served as U.S. Deputy Chief Technology Officer, helping start the U.S. Digital Services under the second Obama administration, and as a member of the Defense Innovation Network.   Read Jennifer’s book Recoding America and check out her Substack Eating Policy.   Jennifer’s recommended reading: Hack Your Bureaucracy by Marina Nitze & Nick Sinai Crisis Engineering by Marina Nitze, Matthew Weaver, & Mikey Dickerson The Procedure Fetish by Nicholas Bagley Why Nothing Works by Marc J. Dunkelman Kill It with Fire by Marianne Bellotti

This episode cross-posts a fireside chat with the Ambassadors of India and France to the United States, Amb. Vinay Kwatra and Amb. Laurent Bili. The discussion was recorded at the Wadhwani AI Center’s January 30 conference, “Exploring Global AI Policy Priorities Ahead of the India AI Impact Summit.” A full recording of the conference, including additional panels and speakers, can be found here.

In this episode, we discuss and evaluate the BIS' new export policy for Nvidia's H200 chips (00:31) before turning to Beijing's decision to block H200 imports (20:18). We then unpack the Pentagon's recently published AI Strategy, including the shift it represents in DOW's approach to AI integration (29:17).   Read the CNAS commentary "Unpacking the H200 Export Policy" here.

In this episode, we examine Grok’s public posting of child sexual abuse material and non-consensual intimate imagery (00:27), the legal consequences xAI may face (12:41), and the international policy community's response (19:05). We then unpack New York’s RAISE Act, including the politics leading up to Gov. Hochul’s signature (22:51) and the final outcome of negotiations (28:16).

In this episode, we're joined by Chris McGuire for a conversation about AI and semiconductor export controls. We begin by discussing Chris' career path into AI and national security (1:55), then turn to his views on recent developments, including reports about a Chinese EUV prototype (11:07). We spend the rest of the episode rating common arguments against AI export controls as fact, fiction, or somewhere in-between (40:25).   Chris is a Senior Fellow for China and Emerging Technologies at the Council on Foreign Relations and a leading expert on U.S.-China AI competition. Before joining CFR, he served as a career government official for over a decade, including as Deputy Senior Director for Technology and National Security at the National Security Council (NSC) from 2022 to 2024. Links to some of Chris' recent work, as discussed in the podcast, are included below.   China’s AI Chip Deficit: Why Huawei Can’t Catch Nvidia and U.S. Export Controls Should Remain   Testimony on Strengthening Export Controls on Semiconductor Manufacturing Equipment

Since 2023, a series of global AI summits has brought together world leaders to advance international dialogue and cooperation on artificial intelligence. Building on this momentum, Prime Minister Narendra Modi announced the India AI Impact Summit, which will take place in New Delhi in February 2026. As the first summit in the series to be hosted in a Global South country, the AI Impact Summit aims to amplify Global South perspectives and advance concrete action to address both the opportunities and risks of AI. On December 8, 2025, the CSIS Wadhwani AI Center hosted S. Krishnan, Secretary of India’s Ministry of Electronics and Information Technology (MeitY), for a livestreamed fireside chat with Wadhwani AI Center Senior Adviser Gregory C. Allen. Secretary Krishnan, who leads India’s national AI strategy, will outline India’s policy priorities and share insights into the goals and global aspirations shaping the upcoming AI Impact Summit. He will also offer a comprehensive look at the central role MeitY plays in driving innovation across India’s AI ecosystem. Secretary Krishnan brings more than 35 years of experience in public service, having joined the Indian Administrative Service in 1989. Prior to his current role, he served as the Additional Chief Secretary of the Industries, Investment Promotion and Commerce Department in the Government of Tamil Nadu. He has also served as Senior Advisor in the Office of the Executive Director for India, Sri Lanka, Bangladesh, and Bhutan at the International Monetary Fund, and has represented India in the G20 Expert Groups on International Financial Architecture and Global Financial Safety Nets. Secretary Krishnan holds a bachelor’s degree from St. Stephen’s College in Delhi.

In this episode, we unpack President Trump’s new executive order targeting state AI laws, including how the final version compares to an earlier draft (1:26), and the legal and political challenges it is likely to face (14:46). We then discuss recent Reuters reporting on Meta’s reliance on scam-driven ad revenue (22:12) and what the social media experience suggests about the risks of failing to regulate AI (45:21).

In this episode, we break down the White House’s decision to let Nvidia’s H200 chips be exported to China and Greg’s case against the move (00:33). We then discuss Trump’s planned “One Rule” executive order to preempt state AI laws (18:59), examine the NDAA's proposed AI Futures Steering Committee (23:09), and analyze the Genesis Mission executive order (26:07), comparing its ambitions and funding reality to the Manhattan Project and Apollo program. We close by looking at why major insurers are seeking to exclude AI risks from corporate policies and how that could impact AI adoption, regulation, and governance (40:29).

In this episode, we start by discussing Greg's trip to India and the upcoming India AI Impact Summit in February 2026 (00:29). We then unpack the Trump Administration’s draft executive order to preempt state AI laws (07:46) and break down the European Commission’s new “digital omnibus” package, including proposed adjustments to the AI Act and broader regulatory simplification efforts (17:51). Finally, we discuss Anthropic’s report on a China-backed “highly sophisticated cyber espionage campaign" using Claude and the mixed reactions from cybersecurity and AI policy experts (37:37).

In this episode, Georgia Adamson and Saif Khan from the Institute for Progress join Greg to unpack their October 25 paper, "Should the US Sell Blackwell Chips to China?" They discuss the geopolitical context of the paper (3:26), how the rumored B30A would compare to other advanced AI chips (11:37), and the potential consequences if the US were to permit B30A exports to China (32:00). Their paper is available here.

One of the most common questions we get from listeners is how to build a successful career in AI policy—so we dedicated an entire episode to answering it. We cover the most formative experiences from Greg's career journey (3:30), general principles for professional success (45:09), and actionable tips specific to breaking into the AI policy space (1:11:52).

In this episode, we cover OpenAI’s latest video-generation model Sora 2 (1:02), concrete harms and potential risks from deepfakes (5:18), the underlying technology and its history (27:03), and how policy can mitigate harms (36:31).

 In this episode, we are joined by Rep. Jay Obernolte, one of Congress’s leading voices on AI policy. We discuss his path from developing video games to serving in Congress (00:49), the work of the bipartisan House Task Force on AI and its final report (9:39), competing approaches to designing AI regulation in Congress (16:38), and prospects for federal preemption of state AI legislation (40:32). Congressman Obernolte has represented California’s 23rd district since 2021. He co-chaired the bipartisan House Task Force on AI, leading the development of an extensive December 2024 report outlining a congressional agenda for AI. He also serves as vice-chair of the Congressional AI Caucus and is the only current member of Congress with an advanced degree in Artificial Intelligence, which he earned from UCLA in 1997. Rep. Obernolte previously served in the California State Legislature.

In this episode, we are joined by economist Harry Holzer to discuss how AI is set to transform labor. Holzer was Chief Economist at the U.S. Department of Labor during the Clinton administration and is currently a Professor of Public Policy at Georgetown University. We break down the fundamentals of the labor market (4:00) and the current and future impact of AI automation (10:30). Holzer also reacts to Anthropic CEO Dario Amodei's warning that AI could eliminate half of entry-level white-collar jobs (23:32) and explains why we need better data capturing AI's impact on the labor market (52:53). Harry Holzer recently co-authored a white paper titled "Proactively Developing & Assisting the Workforce in the Age of AI," which is available here.

In this episode, we dive into California's new AI transparency law, SB 53. We explore the bill's history (00:30), contrast it with the more controversial SB 1047 (6:43), break down the specific disclosure requirements for AI labs of different scales (13:38), and discuss how industry stakeholders and policy experts have responded to the legislation (29:47).

In this episode, we're joined by Joseph Majkut, Director of CSIS' Energy Security and Climate Change Program, to take an in-depth look at energy's role in AI. We explore the current state of the U.S. electrical grid (11:34), bottlenecks in the AI data center buildout (43:45), how U.S. energy efforts compare internationally (1:16:06), and more. Joseph has co-authored three reports on AI and energy: AI for the Grid: Opportunities, Risks, and Safeguards (September 2025), The Electricity Supply Bottleneck on U.S. AI Dominance (March 2025), and The AI Power Surge: Growth Scenarios for GenAI Datacenters Through 2030 (March 2025).

In this episode, we discuss how today’s massive AI infrastructure investments compare to the Manhattan Project (00:33), China’s reported ban on Nvidia chips and its implications for export control policy (13:41), Anthropic’s $1.5 billion copyright settlement with authors (33:49), and recent multibillion-dollar AI investments by Nvidia and ASML (44:42).

In this episode, we discuss China's focus on AI adoption (00:58), the underlying factors driving investor enthusiasm (14:51), and the national security implications of China's booming AI industry (31:47).

In this episode, we are joined by Marietje Schaake, former Member of the European Parliament, to unpack the EU AI Act Code of Practice. Schaake served as Chair of the Working Group on Internal Risk Management and Governance of General-Purpose AI Providers for the Code of Practice, with a focus on AI model safety and security. We discuss the development and drafting of the EU AI Act and Code of Practice (16:47), break down how the Code helps AI companies demonstrate compliance with the Act (28:25), and explore the kinds of systemic risks the AI Act seeks to address (32:00).

In this episode, we unpack the Trump administration’s $8.9 billion deal to acquire a 9.9% stake in Intel, examining the underlying logic, financial terms, and political reactions from across the spectrum (00:33). We then cover Nvidia’s sudden halt in H20 chip production for China, its plans for a Blackwell alternative, and what Beijing’s self-sufficiency push means for the AI race (28:18).