AI Frontiers

<p> Subtitle: Society is not prepared for a flood of agents. We need new protocols and standards, such as Agent ID, to make agents accountable to our legal and financial systems.</p> <p> Gillian Hadfield, Professor of AI Alignment at Johns Hopkins University, Dan Hendrycks, Director of the Center for AI Safety, and Leo Wu, Program Manager at the Center for AI Safety — August 27, 2026</p><p> Last month, Cloudflare reported that more than 50% of internet traffic is now non-human, including a 1,700%+ increase in requests from AI agents. This statistic illustrates an ongoing proliferation of agents—AIs that act autonomously—into the world, taking actions alongside humans. This influx could have both positive and negative effects. While McKinsey predicts that AI agents could create 2.9 trillion dollars in economic value by 2030 in the US alone, the recent cyberattack on Hugging Face, conducted autonomously by OpenAI agents, demonstrates one of the many risks of misaligned AI agents. Meanwhile, the speed at which agents can learn certain skills has been doubling every 3 months.</p><p> As these autonomous agents are introduced into our economies and societies, we will need new infrastructure—laws, protocols, and institutions—to make autonomous AIs accountable to existing legal and financial [...]</p> <p>---</p><p><strong>Outline:</strong></p><p>(03:07) IDs Make Agents Accountable to Governance</p><p>(05:40) A Concrete Agent ID Proposal</p><p>(09:53) Model Deployment Cards</p><p>(13:02) Agent Personhood</p><p>(15:49) Agents and Payment Systems</p><p>(18:47) Agent Infrastructure Is Central to AI Governance</p> <p>---</p> <p><b>First published:</b><br/> August 27th, 2026 </p> <p><b>Source:</b><br/> <a href="https://newsletter.ai-frontiers.org/p/we-need-better-infrastructure-to?utm_source=TYPE_III_AUDIO&utm_medium=Podcast&utm_content=Source+URL+in+episode+description&utm_campaign=ai_narration" rel="noopener noreferrer" target="_blank">https://newsletter.ai-frontiers.org/p/we-need-better-infrastructure-to</a> </p> <p>---</p> <p>Narrated by <a href="https://type3.audio/?utm_source=TYPE_III_AUDIO&utm_medium=Podcast&utm_content=Narrated+by+TYPE+III+AUDIO&utm_term=center_for_ai_safety&utm_campaign=ai_narration" rel="noopener noreferrer" target="_blank">TYPE III AUDIO</a>.</p> <p>---</p><div style="max-width: 100%";><p><strong>Images from the article:</strong></p><a href="https://substackcdn.com/image/fetch/$s_!_OTO!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Fd325a6a0-fe26-424d-aaf2-0966503f2221_7500x4000.jpeg" target="_blank"><img src="https://substackcdn.com/image/fetch/$s_!_OTO!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Fd325a6a0-fe26-424d-aaf2-0966503f2221_7500x4000.jpeg" alt="Interlocking mechanical gears illuminated in orange and blue lighting." style="max-width: 100%;" /></a><hr style="margin-top: 24px; margin-bottom: 24px;" /><a href="https://substackcdn.com/image/fetch/$s_!edbI!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F9ae61d14-4506-40d1-b65a-e42322a22ebd_2048x752.png" target="_blank"><img src="https://substackcdn.com/image/fetch/$s_!edbI!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F9ae61d14-4506-40d1-b65a-e42322a22ebd_2048x752.png" alt="Agents provide information to trusted registries and receive a PPID, which they can present to counterparties they are transacting with. Counterparties can verify the validity of a PPID, and potentially query the registry for additional information. The registry otherwise keeps the connection between PPIDs and underlying profiles private. But if the agent causes harm, counterparties and third parties can submit legal requests to deanonymize the agent and principal." style="max-width: 100%;" /></a><hr style="margin-top: 24px; margin-bottom: 24px;" /><a href="https://substackcdn.com/image/fetch/$s_!HcVp!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F84f8ca1e-3de9-4708-90d5-02ae78146e24_2048x1094.jpeg" target="_blank"><img src="https://substackcdn.com/image/fetch/$s_!HcVp!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F84f8ca1e-3de9-4708-90d5-02ae78146e24_2048x1094.jpeg" alt="Model deployment cards measure the real-world impacts of AI agents. Reported metrics can be split into two categories: internal deployment metrics—how models are used and observed inside AI companies—and external deployment metrics—how models are used throughout the world. AI companies might disclose additional, sensitive metrics in versions of the report available to auditors and regulators." style="max-width: 100%;" /></a><p><em>Apple Podcasts and Spotify do not show images in the episode description. Try <a href="https://pocketcasts.com/" target="_blank" rel="noreferrer">Pocket Casts</a>, or another podcast app.</em></p></div>

<p> Subtitle: For years, the federal government has required insurers to price hypothetical terrorist attacks. It should run the same exercise for the most severe AI risks.</p> <p> Daniel Carpenter, Professor of Government at Harvard University, Feodora Douplitzky-Lunati, and Arjun Purohit — August 24, 2026</p><p> In April, US Treasury Secretary Scott Bessent and then–Federal Reserve Chair Jerome Powell gathered leading banks to discuss the threats of Anthropic's new AI model, Claude Mythos. It was a tacit admission: no one, in the public or private sector, is ready for what is coming. The Mythos announcement led top AI firms to attach tighter restrictions to their own models and prompted even relatively laissez-faire voices in the federal government and think tanks to begin considering preapproval regimes for frontier AI models. Wherever those debates go now, Mythos has changed the game, bringing into sharp focus the vast potential scope and scale of catastrophic risks posed by AI—such as an AI-triggered financial meltdown, vast cybersecurity or infrastructure collapse, or enhanced terrorism risk. As a result, companies, the government, engineers, scholars, and all of society are beginning to look for a different set of institutions, whether in regulation or insurance, to manage the risks.</p><p> However [...]</p> <p>---</p><p><strong>Outline:</strong></p><p>(03:16) Analyzing and Insuring Against Novel Threats</p><p>(09:15) How TRIP Data Calls Benefit the Public</p><p>(16:01) Can the TRIP Model Transfer and Scale?</p><p>(21:36) Discussion about this post</p><p>(21:40) Ready for more?</p> <p>---</p> <p><b>First published:</b><br/> August 24th, 2026 </p> <p><b>Source:</b><br/> <a href="https://newsletter.ai-frontiers.org/p/we-dont-need-to-wait-for-an-ai-disaster?utm_source=TYPE_III_AUDIO&utm_medium=Podcast&utm_content=Source+URL+in+episode+description&utm_campaign=ai_narration" rel="noopener noreferrer" target="_blank">https://newsletter.ai-frontiers.org/p/we-dont-need-to-wait-for-an-ai-disaster</a> </p> <p>---</p> <p>Narrated by <a href="https://type3.audio/?utm_source=TYPE_III_AUDIO&utm_medium=Podcast&utm_content=Narrated+by+TYPE+III+AUDIO&utm_term=center_for_ai_safety&utm_campaign=ai_narration" rel="noopener noreferrer" target="_blank">TYPE III AUDIO</a>.</p> <p>---</p><div style="max-width: 100%";><p><strong>Images from the article:</strong></p><a href="https://substackcdn.com/image/fetch/$s_!0qkW!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Fbf76c04b-9440-45e7-a268-e16878ab6afb_7000x2800.jpeg" target="_blank"><img src="https://substackcdn.com/image/fetch/$s_!0qkW!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Fbf76c04b-9440-45e7-a268-e16878ab6afb_7000x2800.jpeg" alt="Yellow caution tape reading "DO NOT ENTER."" style="max-width: 100%;" /></a><hr style="margin-top: 24px; margin-bottom: 24px;" /><a href="https://substackcdn.com/image/fetch/$s_!O5T7!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Fd1df00e9-e103-44f1-94bd-0dc119f6bc85_2048x1811.png" target="_blank"><img src="https://substackcdn.com/image/fetch/$s_!O5T7!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Fd1df00e9-e103-44f1-94bd-0dc119f6bc85_2048x1811.png" alt="Outline of the TRIP data call process." style="max-width: 100%;" /></a><p><em>Apple Podcasts and Spotify do not show images in the episode description. Try <a href="https://pocketcasts.com/" target="_blank" rel="noreferrer">Pocket Casts</a>, or another podcast app.</em></p></div>

<p> Subtitle: As AI makes legal work significantly cheaper, the burden on courts may increase. Some areas of law could see precedent shift faster as a result.</p> <p> Henry Thompson, Professor of Economics at the University of Mississippi — August 17, 2026</p><p> History shows that new technologies have often brought about significant changes in the law. The proliferation of rail and heavy machinery during the Industrial Revolution increased the number and severity of accidents. This spurred a shift from strict liability (businesses were liable for harm caused by their activities) to negligence (businesses would be held responsible if they failed to take reasonable care) in cases involving railroad and industrial accidents. Then, in the 20th century, as consumer goods became more complex and supply chains grew, it became much harder for consumers to prove negligence on the part of the producer if a product was faulty. This drove the rise of strict products liability, shifting responsibility for defective products toward manufacturers.</p><p> Today, the emergence of AI offers a rare opportunity to analyze how the law will change in response to an altogether different kind of innovation: a labor-augmenting technology that is especially likely to change lawyer productivity.</p><p> In a recent [...]</p> <p>---</p><p><strong>Outline:</strong></p><p>(02:14) AI's Effects on the Practice of Law</p><p>(05:08) The Number of Disputes</p><p>(09:51) The Settlement Rate</p><p>(11:53) AI and the Evolution of Law</p><p>(14:21) Hyperlaw in Society</p><p>(17:09) Discussion about this post</p><p>(17:13) Ready for more?</p> <p>---</p> <p><b>First published:</b><br/> August 17th, 2026 </p> <p><b>Source:</b><br/> <a href="https://newsletter.ai-frontiers.org/p/hyperlaw-ai-will-change-how-law-evolves?utm_source=TYPE_III_AUDIO&utm_medium=Podcast&utm_content=Source+URL+in+episode+description&utm_campaign=ai_narration" rel="noopener noreferrer" target="_blank">https://newsletter.ai-frontiers.org/p/hyperlaw-ai-will-change-how-law-evolves</a> </p> <p>---</p> <p>Narrated by <a href="https://type3.audio/?utm_source=TYPE_III_AUDIO&utm_medium=Podcast&utm_content=Narrated+by+TYPE+III+AUDIO&utm_term=center_for_ai_safety&utm_campaign=ai_narration" rel="noopener noreferrer" target="_blank">TYPE III AUDIO</a>.</p> <p>---</p><div style="max-width: 100%";><p><strong>Images from the article:</strong></p><a href="https://substackcdn.com/image/fetch/$s_!Hldy!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F8ef9fa9f-9315-45a9-9ca1-131e1275833a_5000x2000.jpeg" target="_blank"><img src="https://substackcdn.com/image/fetch/$s_!Hldy!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F8ef9fa9f-9315-45a9-9ca1-131e1275833a_5000x2000.jpeg" alt="Bronze Lady Justice statue holding scales against orange background." style="max-width: 100%;" /></a><hr style="margin-top: 24px; margin-bottom: 24px;" /><a href="https://substackcdn.com/image/fetch/$s_!4-eF!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Fb9bc908b-b663-42a4-90ef-1eb1e9bac3b1_2048x1099.png" target="_blank"><img src="https://substackcdn.com/image/fetch/$s_!4-eF!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Fb9bc908b-b663-42a4-90ef-1eb1e9bac3b1_2048x1099.png" alt="AI will reduce both litigation and gap-filling costs, shifting the cost-benefit analysis of filling additional contract gaps. If one cost falls by a greater percentage than the other, incentives will shift toward the activity whose costs fall more." style="max-width: 100%;" /></a><hr style="margin-top: 24px; margin-bottom: 24px;" /><a href="https://substackcdn.com/image/fetch/$s_!fV4H!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F50b1aa99-1430-46a0-8004-fae8e1426fe3_2048x1229.png" target="_blank"><img src="https://substackcdn.com/image/fetch/$s_!fV4H!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F50b1aa99-1430-46a0-8004-fae8e1426fe3_2048x1229.png" alt="Quincy only settles when he receives more than his expected compensation from trial, minus litigation costs. If litigation costs go down, he’d only agree to larger settlements. Similarly, Genevieve would reject expensive settlements if going to trial became cheaper. This would raise the chance both parties fail to settle and go to trial." style="max-width: 100%;" /></a><p><em>Apple Podcasts and Spotify do not show images in the episode description. Try <a href="https://pocketcasts.com/" target="_blank" rel="noreferrer">Pocket Casts</a>, or another podcast app.</em></p></div>

<p> Subtitle: New EU and California laws require AI companies, and eventually camera makers, to sign their media outputs. It's our best chance to tell what's real.</p> <p> Eddan Katz, Head of Policy &amp; Gov Affairs at Encypher — August 13, 2026</p><p> Two days before Slovakia's 2023 parliamentary election, a fabricated audio clip spread widely on social media, appearing to catch a leading candidate and a prominent journalist discussing how to rig the vote. The following year, nonconsensual pornographic deepfakes of Taylor Swift amassed over 45 million views on X in less than 24 hours. In summer 2025, an elderly woman in Ontario lost her life savings to a cryptocurrency scam, which began with a Facebook advertisement featuring a deepfake of Prime Minister Mark Carney.</p><p> In each of these cases, harm might have been reduced by content provenance metadata: information attached to content that declares how it was produced. (Generally, the term “provenance” refers to a record of where something came from and what happened to it along the way.) Content provenance tools can’t directly prove whether a bare image or video is real or synthetic; no technology reliably can. Instead, they serve as an evidentiary layer of AI governance [...]</p> <p>---</p><p><strong>Outline:</strong></p><p>(02:49) What Content Provenance Is and How It Works</p><p>(06:51) Limitations of Content Credentials</p><p>(12:00) What the EU and California Acts Require</p><p>(15:15) The EU and California Laws' Global Impact</p><p>(17:43) How Provenance Mandates Support Courts in Tackling Deepfake Evidence</p><p>(21:21) Discussion about this post</p><p>(21:25) Ready for more?</p> <p>---</p> <p><b>First published:</b><br/> August 13th, 2026 </p> <p><b>Source:</b><br/> <a href="https://newsletter.ai-frontiers.org/p/ai-content-must-now-carry-a-label?utm_source=TYPE_III_AUDIO&utm_medium=Podcast&utm_content=Source+URL+in+episode+description&utm_campaign=ai_narration" rel="noopener noreferrer" target="_blank">https://newsletter.ai-frontiers.org/p/ai-content-must-now-carry-a-label</a> </p> <p>---</p> <p>Narrated by <a href="https://type3.audio/?utm_source=TYPE_III_AUDIO&utm_medium=Podcast&utm_content=Narrated+by+TYPE+III+AUDIO&utm_term=center_for_ai_safety&utm_campaign=ai_narration" rel="noopener noreferrer" target="_blank">TYPE III AUDIO</a>.</p> <p>---</p><div style="max-width: 100%";><p><strong>Images from the article:</strong></p><a href="https://substackcdn.com/image/fetch/$s_!Z3ZI!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Fb33cf680-6249-4b33-9913-e23bcea4d771_5470x2188.jpeg" target="_blank"><img src="https://substackcdn.com/image/fetch/$s_!Z3ZI!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Fb33cf680-6249-4b33-9913-e23bcea4d771_5470x2188.jpeg" alt="Rose gold laptop and camera on gimbal, wooden desk." style="max-width: 100%;" /></a><hr style="margin-top: 24px; margin-bottom: 24px;" /><a href="https://substackcdn.com/image/fetch/$s_!nyRt!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F5a7a69be-a2c1-4090-ae18-fc9329e65ed9_2048x922.png" target="_blank"><img src="https://substackcdn.com/image/fetch/$s_!nyRt!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F5a7a69be-a2c1-4090-ae18-fc9329e65ed9_2048x922.png" alt="A simplified representation of how Content Credentials metadata is generated. The original data, such as a raw image, and statements about it are hashed and combined into a set of numbers called a “claim.” Hardware manufacturers embed trusted keys in their devices to “sign” claims. With Content Credentials, outside observers can easily verify that the data and statements were signed by the hardware manufacturer’s key." style="max-width: 100%;" /></a><hr style="margin-top: 24px; margin-bottom: 24px;" /><a href="https://substackcdn.com/image/fetch/$s_!0lXk!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F08f1c82e-64f9-474f-bbdc-03fdce0170dc_1680x1650.png" target="_blank"><img src="https://substackcdn.com/image/fetch/$s_!0lXk!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F08f1c82e-64f9-474f-bbdc-03fdce0170dc_1680x1650.png" alt="A tool provided by the Content Authenticity Initiative displays C2PA metadata from an image generated in ChatGPT, revealing its true provenance." style="max-width: 100%;" /></a><hr style="margin-top: 24px; margin-bottom: 24px;" /><a href="https://substackcdn.com/image/fetch/$s_!e9Bq!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Fdfbe6df4-4e0b-4140-8bc0-3d9a15648dee_694x727.png" target="_blank"><img src="https://substackcdn.com/image/fetch/$s_!e9Bq!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Fdfbe6df4-4e0b-4140-8bc0-3d9a15648dee_694x727.png" alt="Sony’s Camera Authenticity Solution makes it hard to pass off synthetic images as genuine by photographing them off a screen, addressing the signing-through-camera vulnerability. Source: Sony." style="max-width: 100%;" /></a><p><em>Apple Podcasts and Spotify do not show images in the episode description. Try <a href="https://pocketcasts.com/" target="_blank" rel="noreferrer">Pocket Casts</a>, or another podcast app.</em></p></div>

<p> Subtitle: If AI reaches the point where it can do the cognitive work humans do, robots will proliferate. Standard data on US industry implies a fully automated economy could double its output roughly every year. </p> <p> Damon Binder, Senior Researcher at Coefficient Giving — August 11, 2026</p><p> AI systems now write so fluently that cheating at universities has become ubiquitous and “AI slop” is displacing human writing across the internet. They have begun producing original proofs on long-standing open problems in mathematics. Researchers at the frontier AI companies have largely stopped writing their own code, and the models’ hacking abilities are strong enough that the US government temporarily export-controlled Anthropic's Claude Fable 5, days after its release. Ten years ago, every claim in this paragraph would have sounded like science fiction.</p><p> It is easy to become numb to this progress. Today's AI agents remain endearingly clumsy; watching Claude run a small store is a bit like watching a child running a lemonade stand. But we should not let such awkwardness distract from the trend. These systems have evolved from research curiosities to extremely powerful economic engines in 10 years, with no end in sight. I want to take [...]</p> <p>---</p><p><strong>Outline:</strong></p><p>(03:17) Why AGI Gives You Robots</p><p>(06:50) Machine Labor Will Be Cheap</p><p>(08:20) Input-Output Analysis</p><p>(11:38) How Fast Could an Autonomous Economy Grow?</p><p>(17:03) From Investment Boom to Unprecedented Growth</p><p>(20:50) What Could Stop an AI Industrial Explosion?</p><p>(24:01) Power Without People</p><p>(26:15) Discussion about this post</p><p>(26:19) Ready for more?</p> <p>---</p> <p><b>First published:</b><br/> August 11th, 2026 </p> <p><b>Source:</b><br/> <a href="https://newsletter.ai-frontiers.org/p/agi-will-set-off-an-industrial-explosion?utm_source=TYPE_III_AUDIO&utm_medium=Podcast&utm_content=Source+URL+in+episode+description&utm_campaign=ai_narration" rel="noopener noreferrer" target="_blank">https://newsletter.ai-frontiers.org/p/agi-will-set-off-an-industrial-explosion</a> </p> <p>---</p> <p>Narrated by <a href="https://type3.audio/?utm_source=TYPE_III_AUDIO&utm_medium=Podcast&utm_content=Narrated+by+TYPE+III+AUDIO&utm_term=center_for_ai_safety&utm_campaign=ai_narration" rel="noopener noreferrer" target="_blank">TYPE III AUDIO</a>.</p> <p>---</p><div style="max-width: 100%";><p><strong>Images from the article:</strong></p><a href="https://substackcdn.com/image/fetch/$s_!tU8F!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Ff2c7356f-e943-47a8-87d2-d6ac2f06d10b_4800x2400.jpeg" target="_blank"><img src="https://substackcdn.com/image/fetch/$s_!tU8F!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Ff2c7356f-e943-47a8-87d2-d6ac2f06d10b_4800x2400.jpeg" alt="Yellow robotic arms along an industrial conveyor belt in warehouse." style="max-width: 100%;" /></a><hr style="margin-top: 24px; margin-bottom: 24px;" /><a href="https://substackcdn.com/image/fetch/$s_!xRIM!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Fc85a0e91-eb43-4f12-8f47-7b0228388274_2048x1588.png" target="_blank"><img src="https://substackcdn.com/image/fetch/$s_!xRIM!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Fc85a0e91-eb43-4f12-8f47-7b0228388274_2048x1588.png" alt="Projected physical output of four key industrial sectors in the years after AGI arrives, relative to today’s level. Each scenario assumes that new capacity takes zero, six months, or a year to build before it starts producing output. Solid lines hold human consumption at today’s level and reinvest all output above that level into industry; dashed lines split that output 50-50 between industry and human consumption. A real buildout would differ somewhat from these projections: it would be slower initially as AI capabilities diffuse throughout the economy, but accelerate later as production methods adapt to robot workers. Source: author’s calculations from the 2017 US input-output tables, as described in Part 2 of the author’s memo series." style="max-width: 100%;" /></a><hr style="margin-top: 24px; margin-bottom: 24px;" /><a href="https://substackcdn.com/image/fetch/$s_!D7B4!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Fdc62d4a7-9983-4506-b59f-30d9be472006_2048x1863.png" target="_blank"><img src="https://substackcdn.com/image/fetch/$s_!D7B4!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Fdc62d4a7-9983-4506-b59f-30d9be472006_2048x1863.png" alt="The self-replicating economy at maximum growth, computed from the 2017 US input-output tables at their most detailed level, then aggregated to the government’s 71 summary sectors for display. Columns give each sector’s share of the human workforce being replaced (at today’s staffing levels), of output (valued at 2017 prices), and of energy use (electricity and fuel purchases, in 2017 dollars)." style="max-width: 100%;" /></a><p><em>Apple Podcasts and Spotify do not show images in the episode description. Try <a href="https://pocketcasts.com/" target="_blank" rel="noreferrer">Pocket Casts</a>, or another podcast app.</em></p></div>

<p> Subtitle: Skeptics of an AI slowdown deal with China say we’d need futuristic tech to make it cheat-proof. Instead, the US and China could just give auditors comprehensive access to major AI companies.</p> <p> Felix Choussat, Visiting Policy Researcher at the Center for AI Safety and Adam Khoja, Researcher at the Center for AI Safety — August 5, 2026</p><p> On July 28, over a thousand employees of the world's top AI companies advocated that the US government “support an international effort to develop the technical and governance tools needed to deliberately pace the frontier of automated AI development.” Following months of cybersecurity scares and their own loss of control incidents, both OpenAI and Anthropic officially endorsed the same message, recognizing the danger of blindly accelerating AI development.</p><p> Despite this new urgency, many people argue that coordinating an international slowdown is currently unworkable—including some of the same groups in favor of one. Even if the US slowed down its own AI development, it wouldn’t be able to make sure that China was doing the same, leaving the US no choice but to race.</p><p> These anti-slowdown arguments usually emphasize the technical challenges with designing AI monitoring measures to ensure a slowdown [...]</p> <p>---</p><p><strong>Outline:</strong></p><p>(05:01) Enforcing a Slowdown Through Whole-Lab Inspections</p><p>(09:39) Joint Verification</p><p>(13:47) Benefits of Slowdown</p><p>(16:10) An AI Slowdown Does Not Require New Technology</p><p>(18:29) Discussion about this post</p><p>(18:33) Ready for more?</p> <p>---</p> <p><b>First published:</b><br/> August 5th, 2026 </p> <p><b>Source:</b><br/> <a href="https://newsletter.ai-frontiers.org/p/an-international-ai-slowdown-is-ready?utm_source=TYPE_III_AUDIO&utm_medium=Podcast&utm_content=Source+URL+in+episode+description&utm_campaign=ai_narration" rel="noopener noreferrer" target="_blank">https://newsletter.ai-frontiers.org/p/an-international-ai-slowdown-is-ready</a> </p> <p>---</p> <p>Narrated by <a href="https://type3.audio/?utm_source=TYPE_III_AUDIO&utm_medium=Podcast&utm_content=Narrated+by+TYPE+III+AUDIO&utm_term=center_for_ai_safety&utm_campaign=ai_narration" rel="noopener noreferrer" target="_blank">TYPE III AUDIO</a>.</p> <p>---</p><div style="max-width: 100%";><p><strong>Images from the article:</strong></p><a href="https://substackcdn.com/image/fetch/$s_!pO55!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F19a220b5-6762-4165-9500-ab51a8309c8e_6320x3160.jpeg" target="_blank"><img src="https://substackcdn.com/image/fetch/$s_!pO55!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F19a220b5-6762-4165-9500-ab51a8309c8e_6320x3160.jpeg" alt="Surveillance system monitor displaying multiple security camera feeds." style="max-width: 100%;" /></a><hr style="margin-top: 24px; margin-bottom: 24px;" /><a href="https://substackcdn.com/image/fetch/$s_!ik10!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Fa901bc21-ffd4-44af-a29f-64d3729d401d_2200x1984.jpeg" target="_blank"><img src="https://substackcdn.com/image/fetch/$s_!ik10!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Fa901bc21-ffd4-44af-a29f-64d3729d401d_2200x1984.jpeg" alt="Diagram showing "Whole-Lab Inspection" with inputs feeding On-Site Inspectors." style="max-width: 100%;" /></a><hr style="margin-top: 24px; margin-bottom: 24px;" /><a href="https://substackcdn.com/image/fetch/$s_!BvWP!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Fd6baa89a-5b54-4e16-a395-b3d1c4b8da38_1580x1236.jpeg" target="_blank"><img src="https://substackcdn.com/image/fetch/$s_!BvWP!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Fd6baa89a-5b54-4e16-a395-b3d1c4b8da38_1580x1236.jpeg" alt="A slowdown regime might allow post-training that improves targeted safety metrics, like adversarial robustness, as long as it does not produce a large increase in capabilities. This differential perspective on AI Safety research is discussed in more detail here__T3A_LINK_IN_POST__." style="max-width: 100%;" /></a><p><em>Apple Podcasts and Spotify do not show images in the episode description. Try <a href="https://pocketcasts.com/" target="_blank" rel="noreferrer">Pocket Casts</a>, or another podcast app.</em></p></div>

<p> Subtitle: Researchers who find dangerous flaws in frontier models have nowhere safe to report them. AI needs the disclosure system that cybersecurity built decades ago.</p> <p> Rich Barton-Cooper, Research Manager at MATS and Adam Gleave, CEO of FAR.AI — August 3, 2026</p><p> Frontier AI developers deploy significant safeguards to prevent their AI models from being misused. However, these safeguards are not robust: AI researchers consistently find techniques for circumventing them, commonly known as “jailbreaks.” Once companies learn about a jailbreak, they can usually implement a fix; yet AI researchers currently lack a safe and reliable way to inform AI companies about the jailbreaks they’ve discovered. We propose concrete improvements to the current system based on cybersecurity norms of responsible disclosure.</p><p> Addressing this problem is urgent. Frontier AI models are already being abused by malicious actors: terrorist group Boko Haram is reported to be training its members on jailbreaking techniques in order to troubleshoot weapons, design new explosive devices, and plan attacks. Some former Boko Haram members stated an unequivocal willingness to use chemical or biological weapons—a task jailbroken models may already be able to help with today. With model capabilities continuing to rapidly accelerate—including in offensive cyber, where jailbreaks [...]</p> <p>---</p><p><strong>Outline:</strong></p><p>(04:11) There Is No Good Way to Report Jailbreaks</p><p>(06:12) Current Reporting Mechanisms Are Limited</p><p>(11:19) Fear of Liability May Incentivize Vendors to Ignore Reports</p><p>(12:26) Established Channels Are Often Covered by Powerful NDAs</p><p>(15:11) Labs Grade Jailbreaks Without Oversight</p><p>(16:27) The System Today Looks Increasingly Fragile</p><p>(17:46) Proposals for Better Disclosure Practices</p><p>(24:26) Relevant Parties Should Act Now</p><p>(27:16) Discussion about this post</p><p>(27:21) Ready for more?</p> <p>---</p> <p><b>First published:</b><br/> August 3rd, 2026 </p> <p><b>Source:</b><br/> <a href="https://newsletter.ai-frontiers.org/p/ai-jailbreak-disclosure-is-broken?utm_source=TYPE_III_AUDIO&utm_medium=Podcast&utm_content=Source+URL+in+episode+description&utm_campaign=ai_narration" rel="noopener noreferrer" target="_blank">https://newsletter.ai-frontiers.org/p/ai-jailbreak-disclosure-is-broken</a> </p> <p>---</p> <p>Narrated by <a href="https://type3.audio/?utm_source=TYPE_III_AUDIO&utm_medium=Podcast&utm_content=Narrated+by+TYPE+III+AUDIO&utm_term=center_for_ai_safety&utm_campaign=ai_narration" rel="noopener noreferrer" target="_blank">TYPE III AUDIO</a>.</p> <p>---</p><div style="max-width: 100%";><p><strong>Images from the article:</strong></p><a href="https://substackcdn.com/image/fetch/$s_!solH!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F6681be17-475e-46d5-b069-200871b70d73_3897x1677.jpeg" target="_blank"><img src="https://substackcdn.com/image/fetch/$s_!solH!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F6681be17-475e-46d5-b069-200871b70d73_3897x1677.jpeg" alt="Rows of dark metal mailboxes with numbers and letters." style="max-width: 100%;" /></a><p><em>Apple Podcasts and Spotify do not show images in the episode description. Try <a href="https://pocketcasts.com/" target="_blank" rel="noreferrer">Pocket Casts</a>, or another podcast app.</em></p></div>

<p> Subtitle: Letting AI developers pick their own safety auditors creates a conflict of interest. Requiring liability insurance instead would put insurers’ own capital behind risk assessments.</p> <p> Gabriel Weil, Professor at the University of Houston Law Center — July 29, 2026</p><p> A growing chorus of scholars and policymakers favors letting private organizations—rather than a government regulator—govern frontier AI. In the leading family of proposals, the state sets the safety outcomes it wants and licenses independent verification organizations (IVOs) that compete to certify developers against those outcomes. Gillian Hadfield has developed the idea as “regulatory markets,” in which AI developers must pay for oversight from private regulators that governments license and hold accountable for safety standards. Dean Ball, who likens the arrangement to bank supervision, has argued for a version he calls “private governance,” which a nonprofit named Fathom has converted into model legislation. The rationale is that legislators and agencies are poorly positioned to write good safety rules for frontier AI: they understand these systems less well than the labs building them, and rules fixed in advance cannot keep pace as the technology changes. Private verifiers, meanwhile, are closer to the technology than any agency and are disciplined by [...]</p> <p>---</p><p><strong>Outline:</strong></p><p>(03:12) Building the Right Incentive Structures</p><p>(08:15) A Mandatory Insurance Model for AI Governance</p><p>(12:23) Objections to a Mandatory Insurance Model</p><p>(20:41) Mandatory Insurance Is a Better Form of Private Governance</p><p>(22:38) Discussion about this post</p><p>(22:42) Ready for more?</p> <p>---</p> <p><b>First published:</b><br/> July 29th, 2026 </p> <p><b>Source:</b><br/> <a href="https://newsletter.ai-frontiers.org/p/dont-let-ai-developers-hire-their?utm_source=TYPE_III_AUDIO&utm_medium=Podcast&utm_content=Source+URL+in+episode+description&utm_campaign=ai_narration" rel="noopener noreferrer" target="_blank">https://newsletter.ai-frontiers.org/p/dont-let-ai-developers-hire-their</a> </p> <p>---</p> <p>Narrated by <a href="https://type3.audio/?utm_source=TYPE_III_AUDIO&utm_medium=Podcast&utm_content=Narrated+by+TYPE+III+AUDIO&utm_term=center_for_ai_safety&utm_campaign=ai_narration" rel="noopener noreferrer" target="_blank">TYPE III AUDIO</a>.</p> <p>---</p><div style="max-width: 100%";><p><strong>Images from the article:</strong></p><a href="https://substackcdn.com/image/fetch/$s_!d0lK!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F332a9a87-95fa-490f-83d1-9c8564d97b4e_8000x3546.jpeg" target="_blank"><img src="https://substackcdn.com/image/fetch/$s_!d0lK!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F332a9a87-95fa-490f-83d1-9c8564d97b4e_8000x3546.jpeg" alt="Gavel, cash, and wooden figure on green background." style="max-width: 100%;" /></a><hr style="margin-top: 24px; margin-bottom: 24px;" /><a href="https://substackcdn.com/image/fetch/$s_!Jc_1!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F80800aa9-b591-4fe8-b765-dd58ef5745b0_2048x1921.png" target="_blank"><img src="https://substackcdn.com/image/fetch/$s_!Jc_1!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F80800aa9-b591-4fe8-b765-dd58ef5745b0_2048x1921.png" alt="Flowchart titled "Model 1: Developer-Paid Audits" showing audit relationships and liability shield." style="max-width: 100%;" /></a><hr style="margin-top: 24px; margin-bottom: 24px;" /><a href="https://substackcdn.com/image/fetch/$s_!hfP5!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F5dcc74c9-41bc-42e3-9426-f152ddd1a35d_2048x1921.png" target="_blank"><img src="https://substackcdn.com/image/fetch/$s_!hfP5!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F5dcc74c9-41bc-42e3-9426-f152ddd1a35d_2048x1921.png" alt="Flowchart diagram titled "Model 2: Mandatory Insurance" showing insurance relationships between parties." style="max-width: 100%;" /></a><p><em>Apple Podcasts and Spotify do not show images in the episode description. Try <a href="https://pocketcasts.com/" target="_blank" rel="noreferrer">Pocket Casts</a>, or another podcast app.</em></p></div>

<p> Subtitle: Today's drones can already navigate indoors, track down humans, and deliver a lethal payload. Attackers willing to kill indiscriminately don’t need to wait for much else.</p> <p> Felix Choussat — July 20, 2026</p><p> Drones are cheap, disposable, and the future of war. Over the past four years, we have seen platforms, missiles, and heavy infantry become increasingly obsolete in the face of $500 drones carrying a pack of explosives—a cost advantage that has let Iranians and Ukrainians alike neuter the conventional capabilities of their great power rivals. Eighty percent of casualties in the bloodiest war since 1945 are from drone strikes, Russia has managed to lose one-third of its fleet to a country without a navy, and the US is spending millions of dollars to intercept five-figure Shaheds flying over the Strait of Hormuz.</p><p> All this is the result of a technology that is still immature. The violence inflicted by today's drones is the handiwork of the scant few that manage to evade countermeasures (a mix of radio jamming, high-power microwave weapons, missiles, automatic cannons, interceptor drones, and nets) before making contact. These defenses exploit the inherent limitations of drones—human guidance, GPS feedback, flight exposure, radio links, range—to [...]</p> <p>---</p><p><strong>Outline:</strong></p><p>(03:02) Breaking the Last Barriers to Autonomous Weapons</p><p>(08:38) Future Methods of Weaponization</p><p>(10:52) What an Urban Drone Attack Might Look Like</p><p>(13:26) Defenses Against Drone WMDs May Be Inadequate</p><p>(17:46) Strategic Applications for Rogue Actors</p><p>(21:37) Taking Drone WMDs Seriously</p><p>(23:43) Discussion about this post</p><p>(23:47) Ready for more?</p> <p>---</p> <p><b>First published:</b><br/> July 20th, 2026 </p> <p><b>Last updated:</b><br/> July 27th, 2026 </p> <p><b>Source:</b><br/> <a href="https://newsletter.ai-frontiers.org/p/drone-wmds-dont-need-any-new-technology?utm_source=TYPE_III_AUDIO&utm_medium=Podcast&utm_content=Source+URL+in+episode+description&utm_campaign=ai_narration" rel="noopener noreferrer" target="_blank">https://newsletter.ai-frontiers.org/p/drone-wmds-dont-need-any-new-technology</a> </p> <p>---</p> <p>Narrated by <a href="https://type3.audio/?utm_source=TYPE_III_AUDIO&utm_medium=Podcast&utm_content=Narrated+by+TYPE+III+AUDIO&utm_term=center_for_ai_safety&utm_campaign=ai_narration" rel="noopener noreferrer" target="_blank">TYPE III AUDIO</a>.</p> <p>---</p><div style="max-width: 100%";><p><strong>Images from the article:</strong></p><a href="https://substackcdn.com/image/fetch/$s_!tw1S!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F5b1bd151-ab54-4509-b176-fc9a6e2646bf_6240x4160.jpeg" target="_blank"><img src="https://substackcdn.com/image/fetch/$s_!tw1S!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F5b1bd151-ab54-4509-b176-fc9a6e2646bf_6240x4160.jpeg" alt="Rows of small racing drones lined up on floor." style="max-width: 100%;" /></a><hr style="margin-top: 24px; margin-bottom: 24px;" /><a href="https://substackcdn.com/image/fetch/$s_!_RB2!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F824b34f1-2e4c-4717-b3d7-08c5dd5e40a2_1480x890.png" target="_blank"><img src="https://substackcdn.com/image/fetch/$s_!_RB2!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F824b34f1-2e4c-4717-b3d7-08c5dd5e40a2_1480x890.png" alt="Interior navigation and mapping from an autonomous human-reconnaissance drone. Source: ShieldAI." style="max-width: 100%;" /></a><hr style="margin-top: 24px; margin-bottom: 24px;" /><a href="https://substackcdn.com/image/fetch/$s_!H66p!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F11b4ebc9-8d6a-4140-904e-7eca74362b67_2048x1028.png" target="_blank"><img src="https://substackcdn.com/image/fetch/$s_!H66p!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F11b4ebc9-8d6a-4140-904e-7eca74362b67_2048x1028.png" alt="" style="max-width: 100%;" /></a><p><em>Apple Podcasts and Spotify do not show images in the episode description. Try <a href="https://pocketcasts.com/" target="_blank" rel="noreferrer">Pocket Casts</a>, or another podcast app.</em></p></div>

<p> Drones are cheap, disposable, and the future of war. Over the past four years, we have seen platforms, missiles, and heavy infantry become increasingly obsolete in the face of $500 drones carrying a pack of explosives—a cost advantage that has let Iranians and Ukrainians alike neuter the conventional capabilities of their great power rivals. Eighty percent of casualties in the bloodiest war since 1945 are from drone strikes, Russia has managed to lose one-third of its fleet to a country without a navy, and the US is spending millions of dollars to intercept five-figure Shaheds flying over the Strait of Hormuz.</p><p> All this is the result of a technology that is still immature. The violence inflicted by today's drones is the handiwork of the scant few that manage to evade countermeasures (a mix of radio jamming, high-power microwave weapons, missiles, automatic cannons, interceptor drones, and nets) before making contact. These defenses exploit the inherent limitations of drones—human guidance, GPS feedback, flight exposure, radio links, range—to take them down en masse. And yet, even though 75% of some types of drones manufactured today never reach their targets, they have nonetheless been strategically decisive in Ukraine and elsewhere.</p><p> These limitations will [...]</p> <p>---</p><p><strong>Outline:</strong></p><p>(02:33) Breaking the Last Barriers to Autonomous Weapons</p><p>(08:21) Future Methods of Weaponization</p><p>(10:39) What an Urban Drone Attack Might Look Like</p><p>(13:20) Defenses Against Drone WMDs May Be Inadequate</p><p>(17:46) Strategic Applications for Rogue Actors</p><p>(21:46) Taking Drone WMDs Seriously</p> <p>---</p> <p><b>First published:</b><br/> July 20th, 2026 </p> <p><b>Source:</b><br/> <a href="https://aifrontiersmedia.substack.com/p/drone-wmds-dont-need-any-new-technology?utm_source=TYPE_III_AUDIO&utm_medium=Podcast&utm_content=Source+URL+in+episode+description&utm_campaign=ai_narration" rel="noopener noreferrer" target="_blank">https://aifrontiersmedia.substack.com/p/drone-wmds-dont-need-any-new-technology</a> </p> <p>---</p> <p>Narrated by <a href="https://type3.audio/?utm_source=TYPE_III_AUDIO&utm_medium=Podcast&utm_content=Narrated+by+TYPE+III+AUDIO&utm_term=center_for_ai_safety&utm_campaign=ai_narration" rel="noopener noreferrer" target="_blank">TYPE III AUDIO</a>.</p> <p>---</p><div style="max-width: 100%";><p><strong>Images from the article:</strong></p><a href="https://substackcdn.com/image/fetch/$s_!_RB2!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F824b34f1-2e4c-4717-b3d7-08c5dd5e40a2_1480x890.png" target="_blank"><img src="https://substackcdn.com/image/fetch/$s_!_RB2!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F824b34f1-2e4c-4717-b3d7-08c5dd5e40a2_1480x890.png" alt="Interior navigation and mapping from an autonomous human-reconnaissance drone. Source: ShieldAI." style="max-width: 100%;" /></a><hr style="margin-top: 24px; margin-bottom: 24px;" /><a href="https://substackcdn.com/image/fetch/$s_!H66p!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F11b4ebc9-8d6a-4140-904e-7eca74362b67_2048x1028.png" target="_blank"><img src="https://substackcdn.com/image/fetch/$s_!H66p!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F11b4ebc9-8d6a-4140-904e-7eca74362b67_2048x1028.png" alt="Table comparing delivery systems by hold volume and drone count." style="max-width: 100%;" /></a><p><em>Apple Podcasts and Spotify do not show images in the episode description. Try <a href="https://pocketcasts.com/" target="_blank" rel="noreferrer">Pocket Casts</a>, or another podcast app.</em></p></div>

<p> In September 2025, the state of California made an AI assistant, Poppy, generally available to employees, to help them “explore AI's productivity benefits.” The following month, the North Dakota Legislative Council deployed AI to assist with summarizing bills. And, in March 2026, the Los Angeles Superior Court partnered with an AI company to provide support across several key functions, including “case information, summarization, research, analysis and drafting assistance,” as well as case management.</p><p> In the near future, it is likely that more US states and the federal government will direct the technology toward even more sensitive and significant tasks. AI may be used to adjudicate disputes, accelerate law-enforcement activities, and aid military operations to an even greater extent than it does today. As AI use cases are more regularly documented, scrutinized, fine-tuned, and optimized, people may actually demand increased AI adoption among government actors, even in sensitive domains. That raises the question: which models should be used?</p><p> This question has no simple answer. With other technologies, the choice of a specific model might largely come down to trade-offs between cost and performance. When it comes to AI, however, different frontier models may show significant differences in “character” or appear [...]</p> <p>---</p><p><strong>Outline:</strong></p><p>(02:12) How Model Character Could Influence Policy</p><p>(06:18) Introducing Public Reasoning Fidelity</p><p>(13:18) Designing an Institution to Evaluate PRF</p><p>(16:29) Precedents</p><p>(20:33) The Public Should Choose Public AI Values</p> <p>---</p> <p><b>First published:</b><br/> July 10th, 2026 </p> <p><b>Source:</b><br/> <a href="https://aifrontiersmedia.substack.com/p/the-government-is-choosing-ai-models?utm_source=TYPE_III_AUDIO&utm_medium=Podcast&utm_content=Source+URL+in+episode+description&utm_campaign=ai_narration" rel="noopener noreferrer" target="_blank">https://aifrontiersmedia.substack.com/p/the-government-is-choosing-ai-models</a> </p> <p>---</p> <p>Narrated by <a href="https://type3.audio/?utm_source=TYPE_III_AUDIO&utm_medium=Podcast&utm_content=Narrated+by+TYPE+III+AUDIO&utm_term=center_for_ai_safety&utm_campaign=ai_narration" rel="noopener noreferrer" target="_blank">TYPE III AUDIO</a>.</p>

<p> Subtitle: The public deserves a say over the values of government-procured AIs.</p> <p> Kevin Frazier and Andrew Reddie — July 10, 2026</p><p> In September 2025, the state of California made an AI assistant, Poppy, generally available to employees, to help them “explore AI's productivity benefits.” The following month, the North Dakota Legislative Council deployed AI to assist with summarizing bills. And, in March 2026, the Los Angeles Superior Court partnered with an AI company to provide support across several key functions, including “case information, summarization, research, analysis and drafting assistance,” as well as case management.</p><p> In the near future, it is likely that more US states and the federal government will direct the technology toward even more sensitive and significant tasks. AI may be used to adjudicate disputes, accelerate law-enforcement activities, and aid military operations to an even greater extent than it does today. As AI use cases are more regularly documented, scrutinized, fine-tuned, and optimized, people may actually demand increased AI adoption among government actors, even in sensitive domains. That raises the question: which models should be used?</p><p> This question has no simple answer. With other technologies, the choice of a specific model might largely come down to [...]</p> <p>---</p><p><strong>Outline:</strong></p><p>(02:32) How Model Character Could Influence Policy</p><p>(06:34) Introducing Public Reasoning Fidelity</p><p>(13:23) Designing an Institution to Evaluate PRF</p><p>(16:34) Precedents</p><p>(20:34) The Public Should Choose Public AI Values</p><p>(21:47) Discussion about this post</p><p>(21:51) Ready for more?</p> <p>---</p> <p><b>First published:</b><br/> July 10th, 2026 </p> <p><b>Source:</b><br/> <a href="https://newsletter.ai-frontiers.org/p/the-government-is-choosing-ai-models?utm_source=TYPE_III_AUDIO&utm_medium=Podcast&utm_content=Source+URL+in+episode+description&utm_campaign=ai_narration" rel="noopener noreferrer" target="_blank">https://newsletter.ai-frontiers.org/p/the-government-is-choosing-ai-models</a> </p> <p>---</p> <p>Narrated by <a href="https://type3.audio/?utm_source=TYPE_III_AUDIO&utm_medium=Podcast&utm_content=Narrated+by+TYPE+III+AUDIO&utm_term=center_for_ai_safety&utm_campaign=ai_narration" rel="noopener noreferrer" target="_blank">TYPE III AUDIO</a>.</p> <p>---</p><div style="max-width: 100%";><p><strong>Images from the article:</strong></p><a href="https://substackcdn.com/image/fetch/$s_!eE-7!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Ffac04f95-da68-4756-8fc8-43ee6fde8925_6000x2702.jpeg" target="_blank"><img src="https://substackcdn.com/image/fetch/$s_!eE-7!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Ffac04f95-da68-4756-8fc8-43ee6fde8925_6000x2702.jpeg" alt="Empty jury box with black leather chairs in courtroom." style="max-width: 100%;" /></a><p><em>Apple Podcasts and Spotify do not show images in the episode description. Try <a href="https://pocketcasts.com/" target="_blank" rel="noreferrer">Pocket Casts</a>, or another podcast app.</em></p></div>

<p> Subtitle: One of the most valuable things governments can build today is the capacity to govern advanced AI competently in the future.</p> <p> Charlie Bullock, Senior Research Fellow at the Institute for Law &amp; AI — July 6, 2026</p><p> How should governments regulate the most advanced AI systems? One possible answer is that they should not. Libertarian-minded writers have made the case for a culture of “permissionless innovation” for AI development, in which the role of government would be limited to enforcing existing laws and facilitating industry self-regulation with “soft law” tools such as voluntary standard-setting. Another possibility, more in vogue across the aisle and the pond, would invoke the precautionary principle, using heavy-handed regulation to restrict the development of frontier AI models until developers can adequately prove that their systems are safe.</p><p> I think there's a better approach. In a new essay, my co-author Christoph Winter and I make the case for a governance strategy that we call “radical optionality.” The idea is simple: governments should avoid over-regulation in the short term while building up the institutional capacity needed to competently regulate extremely advanced or “transformative” future AI systems when and if these systems come into existence. The [...]</p> <p>---</p><p><strong>Outline:</strong></p><p>(02:02) Addressing AI Risks Under Uncertainty</p><p>(06:40) Supporting Both Security and Innovation</p><p>(10:37) Discussion about this post</p><p>(10:41) Ready for more?</p> <p>---</p> <p><b>First published:</b><br/> July 6th, 2026 </p> <p><b>Source:</b><br/> <a href="https://newsletter.ai-frontiers.org/p/ai-governance-needs-radical-optionality?utm_source=TYPE_III_AUDIO&utm_medium=Podcast&utm_content=Source+URL+in+episode+description&utm_campaign=ai_narration" rel="noopener noreferrer" target="_blank">https://newsletter.ai-frontiers.org/p/ai-governance-needs-radical-optionality</a> </p> <p>---</p> <p>Narrated by <a href="https://type3.audio/?utm_source=TYPE_III_AUDIO&utm_medium=Podcast&utm_content=Narrated+by+TYPE+III+AUDIO&utm_term=center_for_ai_safety&utm_campaign=ai_narration" rel="noopener noreferrer" target="_blank">TYPE III AUDIO</a>.</p> <p>---</p><div style="max-width: 100%";><p><strong>Images from the article:</strong></p><a href="https://substackcdn.com/image/fetch/$s_!oThs!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Fac394a6c-b2cf-4064-8314-b8ea967e5f67_6000x3500.jpeg" target="_blank"><img src="https://substackcdn.com/image/fetch/$s_!oThs!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Fac394a6c-b2cf-4064-8314-b8ea967e5f67_6000x3500.jpeg" alt="Repeating rows of Earth globes receding into distance." style="max-width: 100%;" /></a><p><em>Apple Podcasts and Spotify do not show images in the episode description. Try <a href="https://pocketcasts.com/" target="_blank" rel="noreferrer">Pocket Casts</a>, or another podcast app.</em></p></div>

<p> How should governments regulate the most advanced AI systems? One possible answer is that they should not. Libertarian-minded writers have made the case for a culture of “permissionless innovation” for AI development, in which the role of government would be limited to enforcing existing laws and facilitating industry self-regulation with “soft law” tools such as voluntary standard-setting. Another possibility, more in vogue across the aisle and the pond, would invoke the precautionary principle, using heavy-handed regulation to restrict the development of frontier AI models until developers can adequately prove that their systems are safe.</p><p> I think there's a better approach. In a new essay, my co-author Christoph Winter and I make the case for a governance strategy that we call “radical optionality.” The idea is simple: governments should avoid over-regulation in the short term while building up the institutional capacity needed to competently regulate extremely advanced or “transformative” future AI systems when and if these systems come into existence. The point of this approach is to maximize optionality by providing our institutions with tools that can be used to respond to a wide range of foreseen or unforeseen future developments.</p><p> In this piece, we explain the rationale behind maximizing [...]</p> <p>---</p><p><strong>Outline:</strong></p><p>(01:40) Addressing AI Risks Under Uncertainty</p><p>(06:26) Supporting Both Security and Innovation</p> <p>---</p> <p><b>First published:</b><br/> July 6th, 2026 </p> <p><b>Source:</b><br/> <a href="https://aifrontiersmedia.substack.com/p/ai-governance-needs-radical-optionality?utm_source=TYPE_III_AUDIO&utm_medium=Podcast&utm_content=Source+URL+in+episode+description&utm_campaign=ai_narration" rel="noopener noreferrer" target="_blank">https://aifrontiersmedia.substack.com/p/ai-governance-needs-radical-optionality</a> </p> <p>---</p> <p>Narrated by <a href="https://type3.audio/?utm_source=TYPE_III_AUDIO&utm_medium=Podcast&utm_content=Narrated+by+TYPE+III+AUDIO&utm_term=center_for_ai_safety&utm_campaign=ai_narration" rel="noopener noreferrer" target="_blank">TYPE III AUDIO</a>.</p>

<p> Subtitle: American leaders agree that the AI race will shape the balance of power with China. But they can’t agree on how to ensure the technology advances American values.</p> <p> Bill Drexel, Senior Fellow at the Hudson Institute — June 30, 2026</p><p> When officials in Washington warn about losing the AI race to China, the conversation turns quickly to military and economic advantage—and rightly so. Advanced AI will reshape everything from weapons systems to medicine, with massive implications for our geopolitical competitiveness. But beneath the great-power competition on AI lies a moral one. The ethical character of the most transformative technology in generations—one that will mediate an ever-larger share of human experience—will be a byproduct of superpower rivalry. At stake is the future of the relationship between individuals and the state, of privacy and control, of human agency and algorithmic authority.</p><p> The fear is not merely that China might build better systems, deploy them more widely, or out-sell American competitors. It is that those systems could carry a tide of new norms shaped by a government that surveils its citizens, suppresses dissent, harbors eugenic ambitions, and treats individual autonomy as a problem. As AI comes to dominate our lives [...]</p> <p>---</p><p><strong>Outline:</strong></p><p>(03:04) Breakaway Tech Dominance</p><p>(07:03) Encoded Values</p><p>(12:59) Emergent Control Regimes</p><p>(18:02) A Rebalanced Approach</p><p>(24:57) Discussion about this post</p><p>(25:01) Ready for more?</p> <p>---</p> <p><b>First published:</b><br/> June 30th, 2026 </p> <p><b>Source:</b><br/> <a href="https://newsletter.ai-frontiers.org/p/three-models-of-sino-american-competition?utm_source=TYPE_III_AUDIO&utm_medium=Podcast&utm_content=Source+URL+in+episode+description&utm_campaign=ai_narration" rel="noopener noreferrer" target="_blank">https://newsletter.ai-frontiers.org/p/three-models-of-sino-american-competition</a> </p> <p>---</p> <p>Narrated by <a href="https://type3.audio/?utm_source=TYPE_III_AUDIO&utm_medium=Podcast&utm_content=Narrated+by+TYPE+III+AUDIO&utm_term=center_for_ai_safety&utm_campaign=ai_narration" rel="noopener noreferrer" target="_blank">TYPE III AUDIO</a>.</p> <p>---</p><div style="max-width: 100%";><p><strong>Images from the article:</strong></p><a href="https://substackcdn.com/image/fetch/$s_!zI0o!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F02b7df73-a6d6-4817-8a56-bc5c1d5e6cee_6014x4014.jpeg" target="_blank"><img src="https://substackcdn.com/image/fetch/$s_!zI0o!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F02b7df73-a6d6-4817-8a56-bc5c1d5e6cee_6014x4014.jpeg" alt="Chinese and American flags on circuit board chips." style="max-width: 100%;" /></a><p><em>Apple Podcasts and Spotify do not show images in the episode description. Try <a href="https://pocketcasts.com/" target="_blank" rel="noreferrer">Pocket Casts</a>, or another podcast app.</em></p></div>

<p> When officials in Washington warn about losing the AI race to China, the conversation turns quickly to military and economic advantage—and rightly so. Advanced AI will reshape everything from weapons systems to medicine, with massive implications for our geopolitical competitiveness. But beneath the great-power competition on AI lies a moral one. The ethical character of the most transformative technology in generations—one that will mediate an ever-larger share of human experience—will be a byproduct of superpower rivalry. At stake is the future of the relationship between individuals and the state, of privacy and control, of human agency and algorithmic authority.</p><p> The fear is not merely that China might build better systems, deploy them more widely, or out-sell American competitors. It is that those systems could carry a tide of new norms shaped by a government that surveils its citizens, suppresses dissent, harbors eugenic ambitions, and treats individual autonomy as a problem. As AI comes to dominate our lives as thoroughly as digital media already has—shaping our health and finances, how our children learn, how we are tracked, even our species’ genetic makeup—this battle for AI's soul will affect us all intimately.</p><p> The contours of that battle are almost always left [...]</p> <p>---</p><p><strong>Outline:</strong></p><p>(02:37) Breakaway Tech Dominance</p><p>(06:45) Encoded Values</p><p>(12:51) Emergent Control Regimes</p><p>(18:05) A Rebalanced Approach</p> <p>---</p> <p><b>First published:</b><br/> June 30th, 2026 </p> <p><b>Source:</b><br/> <a href="https://aifrontiersmedia.substack.com/p/three-models-of-sino-american-competition?utm_source=TYPE_III_AUDIO&utm_medium=Podcast&utm_content=Source+URL+in+episode+description&utm_campaign=ai_narration" rel="noopener noreferrer" target="_blank">https://aifrontiersmedia.substack.com/p/three-models-of-sino-american-competition</a> </p> <p>---</p> <p>Narrated by <a href="https://type3.audio/?utm_source=TYPE_III_AUDIO&utm_medium=Podcast&utm_content=Narrated+by+TYPE+III+AUDIO&utm_term=center_for_ai_safety&utm_campaign=ai_narration" rel="noopener noreferrer" target="_blank">TYPE III AUDIO</a>.</p>

<p> Subtitle: For decades, no nuclear power could disarm its rivals without provoking devastating retaliation. A large AI lead could change that.</p> <p> Govind Pimpale — June 25, 2026</p><p> Since the late 1950s, Mutual Assured Destruction (MAD) has served as a limiting factor on great-power conflict. The doctrine holds that, if two opposing nations have nuclear weapons that can survive one another's initial strike, then the near-certainty of devastating retaliation will deter each side from launching a large-scale nuclear attack. Despite this logic, military planners have long considered the possibility of a counterforce nuclear attack, where a superpower uses nuclear weapons to cripple the nuclear capabilities of its enemy. If such an attack were executed preemptively, as a so-called “first strike,” it could both start and end a great-power conflict in a matter of hours: without retaliatory capacity, the defender would be at the mercy of the aggressor's remaining nuclear weapons and forced to surrender.</p><p> The reason this does not happen is that a truly successful counterforce strike is nearly impossible to pull off: a would-be attacker doesn’t know the locations of all opposing missile launchers and submarines, nor does it have missiles with sufficient precision and speed to destroy [...]</p> <p>---</p><p><strong>Outline:</strong></p><p>(02:47) Why MAD Has Worked So Far</p><p>(06:02) Why AI Could Dramatically Cut Military Infrastructure Costs</p><p>(11:31) What Could We Build?</p><p>(15:22) The Defender's Options</p><p>(17:28) AI May Disrupt Nuclear Deterrence</p><p>(19:04) Discussion about this post</p><p>(19:08) Ready for more?</p> <p><i>The original text contained 1 footnote which was omitted from this narration.</i> </p><p>---</p> <p><b>First published:</b><br/> June 25th, 2026 </p> <p><b>Source:</b><br/> <a href="https://newsletter.ai-frontiers.org/p/an-ai-capabilities-gap-can-endanger?utm_source=TYPE_III_AUDIO&utm_medium=Podcast&utm_content=Source+URL+in+episode+description&utm_campaign=ai_narration" rel="noopener noreferrer" target="_blank">https://newsletter.ai-frontiers.org/p/an-ai-capabilities-gap-can-endanger</a> </p> <p>---</p> <p>Narrated by <a href="https://type3.audio/?utm_source=TYPE_III_AUDIO&utm_medium=Podcast&utm_content=Narrated+by+TYPE+III+AUDIO&utm_term=center_for_ai_safety&utm_campaign=ai_narration" rel="noopener noreferrer" target="_blank">TYPE III AUDIO</a>.</p> <p>---</p><div style="max-width: 100%";><p><strong>Images from the article:</strong></p><a href="https://substackcdn.com/image/fetch/$s_!bTOj!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F40b37b62-effe-455c-8990-2f11af33322d_4029x2685.jpeg" target="_blank"><img src="https://substackcdn.com/image/fetch/$s_!bTOj!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F40b37b62-effe-455c-8990-2f11af33322d_4029x2685.jpeg" alt="Missile launching from ocean with large smoke plume." style="max-width: 100%;" /></a><hr style="margin-top: 24px; margin-bottom: 24px;" /><a href="https://substackcdn.com/image/fetch/$s_!C-Nn!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F6c20260e-3108-4c8a-bffa-4dff0086c412_2430x750.png" target="_blank"><img src="https://substackcdn.com/image/fetch/$s_!C-Nn!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F6c20260e-3108-4c8a-bffa-4dff0086c412_2430x750.png" alt="SEC filing data from Rocket Lab and Firefly Aerospace." style="max-width: 100%;" /></a><p><em>Apple Podcasts and Spotify do not show images in the episode description. Try <a href="https://pocketcasts.com/" target="_blank" rel="noreferrer">Pocket Casts</a>, or another podcast app.</em></p></div>

<p> Since the late 1950s, Mutual Assured Destruction (MAD) has served as a limiting factor on great-power conflict. The doctrine holds that, if two opposing nations have nuclear weapons that can survive one another's initial strike, then the near-certainty of devastating retaliation will deter each side from launching a large-scale nuclear attack. Despite this logic, military planners have long considered the possibility of a counterforce nuclear attack, where a superpower uses nuclear weapons to cripple the nuclear capabilities of its enemy. If such an attack were executed preemptively, as a so-called “first strike,” it could both start and end a great-power conflict in a matter of hours: without retaliatory capacity, the defender would be at the mercy of the aggressor's remaining nuclear weapons and forced to surrender.</p><p> The reason this does not happen is that a truly successful counterforce strike is nearly impossible to pull off: a would-be attacker doesn’t know the locations of all opposing missile launchers and submarines, nor does it have missiles with sufficient precision and speed to destroy opposing missile silos before they could launch retaliatory nuclear warheads. Thus, if provoked by a first strike, the opposing side would likely be able to launch a large-scale [...]</p> <p>---</p><p><strong>Outline:</strong></p><p>(02:26) Why MAD Has Worked So Far</p><p>(05:40) Why AI Could Dramatically Cut Military Infrastructure Costs</p><p>(11:17) What Could We Build?</p><p>(15:06) The Defender's Options</p><p>(17:10) AI May Disrupt Nuclear Deterrence</p> <p><i>The original text contained 1 footnote which was omitted from this narration.</i> </p><p>---</p> <p><b>First published:</b><br/> June 25th, 2026 </p> <p><b>Source:</b><br/> <a href="https://aifrontiersmedia.substack.com/p/an-ai-capabilities-gap-can-endanger?utm_source=TYPE_III_AUDIO&utm_medium=Podcast&utm_content=Source+URL+in+episode+description&utm_campaign=ai_narration" rel="noopener noreferrer" target="_blank">https://aifrontiersmedia.substack.com/p/an-ai-capabilities-gap-can-endanger</a> </p> <p>---</p> <p>Narrated by <a href="https://type3.audio/?utm_source=TYPE_III_AUDIO&utm_medium=Podcast&utm_content=Narrated+by+TYPE+III+AUDIO&utm_term=center_for_ai_safety&utm_campaign=ai_narration" rel="noopener noreferrer" target="_blank">TYPE III AUDIO</a>.</p> <p>---</p><div style="max-width: 100%";><p><strong>Images from the article:</strong></p><a href="https://substackcdn.com/image/fetch/$s_!C-Nn!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F6c20260e-3108-4c8a-bffa-4dff0086c412_2430x750.png" target="_blank"><img src="https://substackcdn.com/image/fetch/$s_!C-Nn!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F6c20260e-3108-4c8a-bffa-4dff0086c412_2430x750.png" alt="SEC filing data from Rocket Lab and Firefly Aerospace." style="max-width: 100%;" /></a><p><em>Apple Podcasts and Spotify do not show images in the episode description. Try <a href="https://pocketcasts.com/" target="_blank" rel="noreferrer">Pocket Casts</a>, or another podcast app.</em></p></div>

<p> On June 12, the Trump administration issued an order requiring Anthropic to suspend access to its two most advanced AI models—Fable 5 and Mythos 5—for non-American nationals, just days after their public release. The decision caused immediate backlash across Europe. Politicians from France's Gabriel Attal and Jordan Bardella to the Netherlands’ Geert Wilders, alongside European industry and civil society, amplified calls for AI sovereignty. Attal, the presidential candidate for Macron's Renaissance party, even likened the shutdown of Anthropic's models to Iran's blockade of the Strait of Hormuz.</p><p> While such reactions are understandable, they underscore what we already know: Europe is behind on AI, overly dependent on the US, and vulnerable to unilateral decisions. Perhaps the renewed sovereignty rhetoric can help build the political capital needed to fix the continent's AI positioning, but what Europe needs more urgently is a clear-eyed account of what to actually do. In the wake of the Anthropic episode, this piece separates what has genuinely changed from what is being overstated, while identifying where European policymakers should focus their attention.</p><p><strong> Europe Should Have Planned for This</strong></p><p> In recent years, AI has become increasingly relevant to national security. Europe could have anticipated that the US government [...]</p> <p>---</p><p><strong>Outline:</strong></p><p>(01:23) Europe Should Have Planned for This</p><p>(03:51) The Restrictions' Short-Term Impact on Europe May Be Limited</p><p>(06:44) In the Long Term, These Restrictions May Provide Opportunities for Europe</p><p>(10:45) How Does Europe Build Leverage When It Needs Results Now?</p><p>(14:27) Export Controls Are a Wake-Up Call for Europe</p> <p>---</p> <p><b>First published:</b><br/> June 19th, 2026 </p> <p><b>Source:</b><br/> <a href="https://aifrontiersmedia.substack.com/p/what-export-controls-on-anthropics?utm_source=TYPE_III_AUDIO&utm_medium=Podcast&utm_content=Source+URL+in+episode+description&utm_campaign=ai_narration" rel="noopener noreferrer" target="_blank">https://aifrontiersmedia.substack.com/p/what-export-controls-on-anthropics</a> </p> <p>---</p> <p>Narrated by <a href="https://type3.audio/?utm_source=TYPE_III_AUDIO&utm_medium=Podcast&utm_content=Narrated+by+TYPE+III+AUDIO&utm_term=center_for_ai_safety&utm_campaign=ai_narration" rel="noopener noreferrer" target="_blank">TYPE III AUDIO</a>.</p>

<p> The debate about AI's future economic impacts often settles into two camps predicting incompatible futures. One camp insists that AI is a normal technology: simply the next in a long line of economic transformations, each increasing productivity while gradually reallocating labor. The other camp warns that AI will become a great displacer: that automation will hollow out the working class within a decade and eventually disempower large swaths of human workers.</p><p> Each side often treats the other's predictions as unserious, and, consequently, policy debates often split along the same tired fault lines: whether we need reskilling or universal basic income, whether we should strengthen safety nets or structurally redesign our economy. The two camps’ forecasts diverge so sharply that it can be hard to see that they do not have to be mutually exclusive.</p><p> Rather, a more useful framing treats these predictions as describing different stages of the same overarching transition rather than as competing accounts of the same moment. From a macro perspective, both narratives will play out roughly sequentially, though those phases may overlap substantially across sectors and timelines.</p><p> In the short term, it seems inevitable that AI will look like an accelerated version of past automation [...]</p> <p>---</p><p><strong>Outline:</strong></p><p>(02:38) Near Term: Managing Economic Shocks</p><p>(05:37) Medium Term: Navigating Reorganization and Divergence</p><p>(10:19) Long Term: Restructuring Economies</p><p>(14:05) Conclusion</p><p>(16:01) Discussion about this post</p><p>(16:05) Ready for more?</p> <p>---</p> <p><b>First published:</b><br/> June 16th, 2026 </p> <p><b>Source:</b><br/> <a href="https://aifrontiersmedia.substack.com/p/a-roadmap-for-the-upcoming-labor?utm_source=TYPE_III_AUDIO&utm_medium=Podcast&utm_content=Source+URL+in+episode+description&utm_campaign=ai_narration" rel="noopener noreferrer" target="_blank">https://aifrontiersmedia.substack.com/p/a-roadmap-for-the-upcoming-labor</a> </p> <p>---</p> <p>Narrated by <a href="https://type3.audio/?utm_source=TYPE_III_AUDIO&utm_medium=Podcast&utm_content=Narrated+by+TYPE+III+AUDIO&utm_term=center_for_ai_safety&utm_campaign=ai_narration" rel="noopener noreferrer" target="_blank">TYPE III AUDIO</a>.</p>