AI Deception, Interpretability, and Affordances with Apollo Research CEO Marius Hobbhahn

Published: Dec. 15, 2023, 8:56 p.m.

In this episode, Marius Hobbhahn, CEO of Apollo Research, sits down with Nathan Labenz to discuss Apollo\u2019s research in AI deception, interpretability, and affordances.\xa0If you need an ecommerce platform, check out our sponsor Shopify: https://shopify.com/cognitive for a $1/month trial period.\n\nWe're hiring across the board at Turpentine and for Erik's personal team on other projects he's incubating. He's hiring a Chief of Staff, EA, Head of Special Projects, Investment Associate, and more. For a list of JDs, check out: eriktorenberg.com.\n\nSPONSORS:\nShopify is the global commerce platform that helps you sell at every stage of your business. Shopify powers 10% of ALL eCommerce in the US. And Shopify's the global force behind Allbirds, Rothy's, and Brooklinen, and 1,000,000s of other entrepreneurs across 175 countries.From their all-in-one e-commerce platform, to their in-person POS system \u2013 wherever and whatever you're selling, Shopify's got you covered. With free Shopify Magic, sell more with less effort by whipping up captivating content that converts \u2013 from blog posts to product descriptions using AI. Sign up for $1/month trial period: https://shopify.com/cognitive\n\nOmneky is an omnichannel creative generation platform that lets you launch hundreds of thousands of ad iterations that actually work customized across all platforms, with a click of a button. Omneky combines generative AI and real-time advertising data. Mention "Cog Rev" for 10% off.\n\nNetSuite has 25 years of providing financial software for all your business needs. More than 36,000 businesses have already upgraded to NetSuite by Oracle, gaining visibility and control over their financials, inventory, HR, eCommerce, and more. If you're looking for an ERP platform \u2705 head to NetSuite: http://netsuite.com/cognitive and download your own customized KPI checklist.\n\nX/SOCIAL:\n@labenz (Nathan)\n@MariusHobbhahn (Marius)\n@eriktorenberg (Erik)\n@CogRev_Podcast\n\nTIMESTAMPS:\n(00:00:00) - Episode Preview\n(00:05:40) - Understanding the Role of Apollo Research\xa0\n(00:08:02) - The Evolution of AI Safety\n(00:11:30) - Apollo\u2019s Framework\n(00:15:10) - Sponsors: Shopify\xa0\n(00:16:49) - Understanding AI Affordances and Resulting Interactions in the World\n(00:31:00) - Sponsors: Omneky\n(00:39:00) - Interpretability and deceptive alignment\xa0\n(00:45:55) - Why might deception arise in the first place?\n(00:47:46) - Understanding deceptive alignment\xa0\n(00:57:49) - New Architectures\n(01:02:23) - Interpretability at deployment phase\n(01:03:35) - Deception in AI | A Case Study\n(01:09:03) - Deceptive AI Stock Trader\n(01:18:51) - Impact of discouragement on unethical AI behaviour\n(01:19:45) - Giving the AI a Reasoning Scratchpad and Impact on Deception\n(01:21:06) - Double-edged sword of removing the scratchpad\n(01:23:00) - Analyzing impact of model size on insider training\xa0\n(01:28:00) - Challenges and necessity of 3rd Party Auditing\n(01:31:00) - Role of government in regulation\xa0\n(01:52:17) - How can individuals get involved in red teaming?\n\nThis show is produced by Turpentine: a network of podcasts, newsletters, and more, covering technology, business, and culture \u2014 all from the perspective of industry insiders and experts. We\u2019re launching new shows every week, and we\u2019re looking for industry-leading sponsors \u2014 if you think that might be you and your company, email us at erik@turpentine.co.