by Jacob Haimes
The Into AI Safety podcast aims to make it easier for everyone, regardless of background, to get meaningfully involved with the conversations surrounding the rules and regulations which should govern the research, development, deployment, and use of the technologies encompassed by the term "artificial intelligence" or "AI" For better formatted show notes, additional resources, and more, go to https://kairos.fm/intoaisafety/
Language
🇺🇲
Publishing Since
11/13/2023
See how your podcast performs across key metrics
Recommendations available
Unlock the full report to see detailed tips
Recommendations available
Unlock the full report to see detailed tips
Unlock comprehensive insights including:
See how your show performs across every key metric
High authority scores make your podcast more attractive to industry leaders and influencers who want to appear on credible shows.
Sponsors look for podcasts with proven authority and engagement. Your score demonstrates your podcast's value to potential partners.
Understanding your strengths and weaknesses helps you make data-driven decisions to expand your listener base effectively.
Pitch yourself as a guest, propose sponsorships, or reach out directly to the host.
July 9, 2026
<p>What if the safest AI models weren't built by adding guardrails after training, but by shaping what gets learned in the first place? Ethan Roland, senior alignment researcher at AE Studio and first author on an ICML 2026 spotlight paper, joins Jacob to talk about gradient routing, a technique that routes dangerous capabilities into isolated parts of a model's architecture where they can be locked or removed entirely. They get into the absorption effect, KYC-style access control frameworks, and what it would actually take for frontier labs to adopt this kind of work before it's needed rather than after. </p><p><strong>Chapters</strong><br></p><ul><li>(00:00) - Introduction </li> <li>(06:39) - Inside AE Studio </li> <li>(15:26) - China & the Alignment vs. Controllability Framing </li> <li>(18:23) - Data Filtering & Gradient Routing (Aside) </li> <li>(30:39) - Mixture of Experts Explained (Aside) </li> <li>(36:25) - Why Pre-Training Interventions Are Rare </li> <li>(42:43) - Ethan's Theory of Change </li> <li>(56:17) - Access Control Governance and KYC (Aside) </li> <li>(01:04:47) - The Researcher's Role in Policy Advocacy </li> <li>(01:11:38) - Speed Round </li> <li>(01:27:55) - Outro</li> </ul><br><strong>Links</strong><br>Below are the most important links for this episode. For more, visit the episode page on <a href="https://kairos.fm/intoaisafety/e032/">Kairos.fm</a>.<ul><li>Ethan's <a href="http://ethan-roland.com/">website</a></li><li>The paper <a href="https://modularpretraining.com/">landing page</a></li><li><a href="https://arxiv.org/abs/2410.04332">Preprint</a> - Gradient Routing: Masking Gradients to Localize Computation in Neural Networks</li><li>ICLR <a href="https://arxiv.org/abs/2508.06601">paper</a> and <a href="https://deepignorance.ai">webpage</a> - Deep Ignorance: Filtering Pretraining Data Builds Tamper-Resistant Safeguards into Open-Weight LLMs</li><li>Wikipedia <a href="https://en.wikipedia.org/wiki/CBRN_defense">article</a> - CBRN defense</li><li>NTI <a href="https://www.nti.org/news/new-nti-tutorials-edcuate-biological-weapons-and-nuclear-testing/">tutorials</a> on bioweapons and nuclear testing</li></ul>
June 1, 2026
<p>Today's episode does double duty as an interview and an announcement. Joe Williams, host of the new Kairos.fm show "Beyond the Paycheck: Reclaiming the Case for UBI in the Age of AI," joins Jacob to talk about his background as a freelance translator and how AI quietly dismantled his livelihood in 2025. From there the conversation expands into whether this moment is really different from past waves of automation, who exactly makes up the tech billionaire class, and why you should probably raise an eyebrow when someone like Elon Musk says he supports UBI. Go check out <a href="https://kairos.fm/reclaimingubi/">Reclaiming UBI</a>!</p><p><strong>Chapters</strong><br></p><ul><li>(00:00) - Intro</li> <li>(05:01) - Joe's Background</li> <li>(19:41) - What is UBI & Why Does It Matter Now?</li> <li>(27:25) - ASIDE: The Corporation Who Cried " My Technology Will End Work"</li> <li>(32:25) - Wealth Concetration & the Digital Economy</li> <li>(39:06) - The AI Gentry</li> <li>(48:13) - ASIDE: Sam Altman + OpenResearch + Worldcoin</li> <li>(53:38) - Powerful UBI Supporters Should Raise Red Flags</li> <li>(01:02:56) - How to Read AI Gentry Discourse Critically</li> <li>(01:10:00) - Who Is "Reclaiming UBI" For?</li> <li>(01:18:17) - Outro</li> </ul><br><strong>Critical Links</strong><br>Below are the most important links for this episode. For more, visit the episode page on <a href="https://kairos.fm/intoaisafety/e031/">Kairos.fm</a>.<ul><li>NEW KAIROS.FM <a href="https://kairos.fm/reclaimingubi">PODCAST</a> - Reclaiming UBI: Work and Values in the Age of AI</li><li>Josh Steimle <a href="https://joshsteimle.com/entrepreneur/ai-destroy-jobs-wrong-predictions.html">blog post</a> - Will AI Destroy Jobs? 500 Years of Predictions Say No</li><li>Smithsonian Magazine <a href="https://www.smithsonianmag.com/history/what-the-luddites-really-fought-against-264412/">article</a> - What the Luddites Really Fought Against</li><li>ITIF <a href="https://itif.org/publications/2022/09/30/oops-the-predicted-47-percent-of-job-loss-from-ai-didnt-happen/">report</a> - Oops: The Predicted 47 Percent of Job Loss From AI Didn't Happen</li><li>Frontiers in Artificial Intelligence <a href="https://www.frontiersin.org/journals/artificial-intelligence/articles/10.3389/frai.2025.1488457/full">journal article</a> - AI, Universal Basic Income, and Power: Symbolic Violence in the Tech Elite's Narrative</li><li>Time <a href="https://time.com/6300522/worldcoin-sam-altman/">article</a> - What to Know About Worldcoin and the Controversy Around It</li><li>Wikipedia <a href="https://en.wikipedia.org/wiki/World_(blockchain">page</a> - World (blockchain)</li></ul>
May 12, 2026
<p>Zainab Majid, co-founder of Asymmetric Security, joins Jacob for a conversation on the intersection between AI Safety and cybersecurity, as well as the future of digital forensics. Drawing from years of incident response work, she explains how cyber attacks actually unfold, why AI is changing both offense and defense, and how her team is building AI-native tools to investigate breaches faster and more effectively. Other topics explored in this episode include trust in the AI/cybersecurity industries, the realities behind cybersecurity hype, and the challenge of keeping humans meaningfully involved as these systems become more capable. Zainab also gives practical, actionable advice on how you can protect yourself.</p><p>If you're interested in over 30 minutes of additional content, head on over to the Kairos.fm <a href="https://www.patreon.com/cw/Kairosfm">Patreon</a> where you can become a subscriber for just $2 per month, which helps make this whole podcasting thing a bit more sustainable.</p><p><strong>Chapters</strong><br></p><ul><li>(00:00) - Intro </li> <li>(04:34) - Zainab's Background </li> <li>(08:49) - Jacob & Zainab's History </li> <li>(16:03) - Founding Asymmetric Security </li> <li>(24:49) - How to Know Who You Can Trust </li> <li>(36:31) - The Threats Asymmetric Is Built to Fight </li> <li>(01:05:54) - What's Asymmetric Tackling Next? </li> <li>(01:15:33) - Glasswing, Dual Use, and Power Concentration </li> <li>(01:24:21) - The Relationship Between AI Safety & Cybersecurity </li> <li>(01:37:01) - Outro</li> </ul><br><strong>Critical Links</strong><br>Below are the most important links for this episode. For more, visit the episode <a href="https://kairos.fm/intoaisafety/e030/">page</a> on Kairos.fm.<ul><li>Asymmetric Security <a href="https://www.asymmetricsecurity.com/">website</a></li><li>EvalEval @ NeurIPS <a href="https://arxiv.org/abs/2411.08813">workshop paper</a> - Rethinking CyberSecEval</li><li>Meta AI <a href="https://arxiv.org/abs/2312.04724">report</a> - Purple Llama CyberSecEval</li><li>Anthropic <a href="https://www.anthropic.com/glasswing">press release</a> - Project Glasswing</li><li>Schneier on Security <a href="https://www.schneier.com/essays/archives/2026/04/what-anthropics-mythos-means-for-the-future-of-cybersecurity.html">blogpost</a> - What Anthropic’s Mythos Means for the Future of Cybersecurity</li></ul>
Guests from recent episodes — sign up to see every guest that has ever appeared on this show.
Li-Lian Ang
Guest
Andres Sepulveda Morales
Guest
Will Petillo
Guest
Tristan Williams
Guest
Felix de Simone
Guest
Dr Igor Krawczuk
Guest
Dr Peter Park
Guest
Have a different question and can't find the answer you're looking for? Reach out to our support team by sending us an email and we'll get back to you as soon as we can.
The Into AI Safety podcast aims to make it easier for everyone, regardless of background, to get meaningfully involved with the conversations surrounding the rules and regulations which should govern the research, development, deployment, and use of the technologies encompassed by the term "artificial intelligence" or "AI"
For better formatted show notes, additional resources, and more, go to https://kairos.fm/intoaisafety/
This podcast updates weekly.
This podcast is available on 8 platforms including Apple Podcasts, Spotify, and more. You can also use the RSS feed directly.
Yes, this podcast regularly features guests.
Pod Engine is not affiliated with, endorsed by, or officially connected with any of the podcasts displayed on this platform. We operate independently as a podcast discovery and analytics service.
All podcast artwork, thumbnails, and content displayed on this page are the property of their respective owners and are protected by applicable copyright laws. This includes, but is not limited to, podcast cover art, episode artwork, show descriptions, episode titles, transcripts, audio snippets, and any other content originating from the podcast creators or their licensors.
We display this content under fair use principles and/or implied license for the purpose of podcast discovery, information, and commentary. We make no claim of ownership over any podcast content, artwork, or related materials shown on this platform. All trademarks, service marks, and trade names are the property of their respective owners.
While we strive to ensure all content usage is properly authorized, if you are a rights holder and believe your content is being used inappropriately or without proper authorization, please contact us immediately at hey@podengine.ai for prompt review and appropriate action, which may include content removal or proper attribution.
By accessing and using this platform, you acknowledge and agree to respect all applicable copyright laws and intellectual property rights of content owners. Any unauthorized reproduction, distribution, or commercial use of the content displayed on this platform is strictly prohibited.