[{"@context":"https:\/\/schema.org\/","@type":"BlogPosting","@id":"https:\/\/blog.terabox.com\/insights\/roman-yampolskiy-ai-safety-simulation-hypothesis#BlogPosting","mainEntityOfPage":"https:\/\/blog.terabox.com\/insights\/roman-yampolskiy-ai-safety-simulation-hypothesis","headline":"Roman Yampolskiy: AI Safety and the Simulation Hypothesis","name":"Roman Yampolskiy: AI Safety and the Simulation Hypothesis","description":"\ud83d\udcfa Today’s recommended deep-dive video: https:\/\/www.youtube.com\/watch?v=TgFmA-Qwsek The Ghost in the Weights: Roman Yampolskiy on the Uncontrollable Rise of SuperintelligenceThe Drive to Win: Defining Intelligence and Self-PreservationThe Matrix Within: Simulation Theory and Hacking RealityThe Final Wall: The Impossibility of AI SafetyKey... ","datePublished":"2026-08-12","dateModified":"2026-08-12","author":{"@type":"Person","@id":"https:\/\/blog.terabox.com\/author\/flextech-admin\/#Person","name":"flextech-admin","url":"https:\/\/blog.terabox.com\/author\/flextech-admin\/","image":{"@type":"ImageObject","@id":"https:\/\/secure.gravatar.com\/avatar\/ad516503a11cd5ca435acc9bb6523536?s=150&d=mm&r=gforcedefault=1","url":"https:\/\/secure.gravatar.com\/avatar\/ad516503a11cd5ca435acc9bb6523536?s=150&d=mm&r=gforcedefault=1","height":96,"width":96}},"publisher":{"@type":"Organization","name":"terabox","logo":{"@type":"ImageObject","@id":"http:\/\/blog.terabox.com\/wp-content\/uploads\/2021\/11\/logo\u4ea7\u54c1\u540d-\u7ad6\u7248.png","url":"http:\/\/blog.terabox.com\/wp-content\/uploads\/2021\/11\/logo\u4ea7\u54c1\u540d-\u7ad6\u7248.png","width":900,"height":900}},"image":{"@type":"ImageObject","@id":"https:\/\/img.youtube.com\/vi\/TgFmA-Qwsek\/maxresdefault.jpg","url":"https:\/\/img.youtube.com\/vi\/TgFmA-Qwsek\/maxresdefault.jpg","height":"","width":""},"url":"https:\/\/blog.terabox.com\/insights\/roman-yampolskiy-ai-safety-simulation-hypothesis","video":{"@context":"http:\/\/schema.org\/","@type":"VideoObject","@id":"https:\/\/www.youtube.com\/watch?v=TgFmA-Qwsek#VideoObject","contentUrl":"https:\/\/www.youtube.com\/watch?v=TgFmA-Qwsek","name":"The Man Behind AI Safety Thinks AI Is Conscious","description":"SPONSORS: \n- Accelerate your efficiency. Sign up for your one-dollar-per-month trial today at http:\/\/shopify.com\/theories\n- I personally subscribe to The Economist. SUMMER SALE to Aug 17, 2026: 50% off the annual subscription! No other podcast has this! https:\/\/economist.com\/TOE\n\nRoman Yampolskiy has spent two decades being right about things people wished he wasn't \u2014 and in this conversation, he's not here to scare you, but to be precise. He makes the case that AI alignment isn't merely unsolved but fundamentally under-defined: no agreed-upon values, no way to formalize them even if there were, and no mechanism for enforcing them on something smarter than its creators. His strongest argument isn't a doom scenario, it's that you cannot indefinitely control something smarter than you.\n\nFOLLOW:\n- Spotify: https:\/\/open.spotify.com\/show\/4gL14b92xAErofYQA7bU4e\n- Substack: https:\/\/curtjaimungal.substack.com\/subscribe\n- Twitter: https:\/\/twitter.com\/TOEwithCurt\n- Discord Invite: https:\/\/discord.com\/invite\/kBcnfNVwqs\n- Crypto: https:\/\/nowpayments.io\/donation\/TOE\n- PayPal: https:\/\/www.paypal.com\/donate?hosted_button_id=XUBHNMFXUX5S4\n\nTIMESTAMPS:\n- 00:00:00 - Defining General Intelligence\n- 00:05:58 - AI Instrumental Convergence\n- 00:11:11 - The Orthogonality Thesis\n- 00:16:15 - Escaping the Simulation\n- 00:21:45 - Principle of Indifference\n- 00:27:51 - Acquired Savant Syndrome\n- 00:33:51 - LLM Internal States\n- 00:41:02 - AI Safety Impossibility Results\n- 00:47:16 - Public Misconceptions\n- 00:53:21 - Existential vs. Suffering Risks\n- 01:01:20 - AI Alignment Definition Crisis\n- 01:09:28 - Computational Irreducibility\n- 01:16:20 - Substrate Independence\n- 01:22:50 - Philosophical Zombie Critique\n- 01:29:57 - The Cassandra Paradox\n- 01:37:35 - Religion and Simulation\n- 01:46:03 - Digital Physics Evidence\n- 01:51:20 - Limits of Control\n\nLINKS MENTIONED:\n- Roman's Papers: https:\/\/scholar.google.com\/citations?user=0_Rq68cAAAAJ\n- Roman's Podcast: https:\/\/www.youtube.com\/channel\/UCPIq6Bb-1iLmqyksJjy4kLQ\n- Roman's Twitter: https:\/\/x.com\/romanyam\n- Roman's Facebook: https:\/\/www.facebook.com\/roman.yampolskiy\n- AI Identity [Paper]: https:\/\/philarchive.org\/archive\/ZIETPO-7\n- Basic AI Drives [Paper]: https:\/\/selfawaresystems.com\/wp-content\/uploads\/2008\/01\/ai_drives_final.pdf\n- Qualia in Agents [Paper]: https:\/\/arxiv.org\/abs\/1712.04020\n- Orthogonality Thesis [Paper]: https:\/\/nickbostrom.com\/superintelligentwill.pdf\n- Escape the Simulation [Paper]: https:\/\/www.researchgate.net\/publication\/369187097_How_to_Escape_From_the_Simulation\n- Could This AI Be Conscious? [Article]: https:\/\/unherd.com\/2026\/05\/is-ai-the-next-phase-of-evolution\n- Impossibility Results in AI [Paper]: https:\/\/arxiv.org\/abs\/2109.00484\n- When AIs Act Emotional: https:\/\/youtu.be\/D4XTefP3Lsc\n- Hacking the Simulation [Paper]: https:\/\/philarchive.org\/rec\/YAMHTS-2\n- Autonomous Machine Intelligence [Paper]: https:\/\/openreview.net\/pdf?id=BZ5a1r-kVsf\n- Hinton on Maternal Instincts [Article]: https:\/\/fortune.com\/2025\/08\/14\/godfather-of-ai-geoffrey-hinton-maternal-instincts-superintelligence\/\n- Singleton Hypothesis [Paper]: https:\/\/nickbostrom.com\/fut\/singleton\n- New Kind of Science [Book]: https:\/\/amazon.com\/dp\/1579550088?tag=toe08-20\n- On AI Controllability [Paper]: https:\/\/arxiv.org\/abs\/2008.04071\n- Universe as Numerical Simulation [Paper]: https:\/\/arxiv.org\/abs\/1210.1847\n- Nir Lahav [TOE]: https:\/\/youtu.be\/3nHiOtnnrzA\n- Joscha Bach [TOE]: https:\/\/youtu.be\/3MNBxfrmfmI\n- Bas Van Fraassen [TOE]: https:\/\/youtu.be\/lhpRAWxvY5s\n- Simulation Hypothesis [TOE]: https:\/\/youtu.be\/3_lBPMc6JRY\n- Geoffrey Hinton [TOE]: https:\/\/youtu.be\/b_DUft-BdIE\n- Max Tegmark [TOE]: https:\/\/youtu.be\/-gekVfUAS7c\n- Stephen Wolfram [TOE]: https:\/\/youtu.be\/FkYer0xP37E\n- David Chalmers [TOE]: https:\/\/youtu.be\/5r9V1ryksnw\n\nMore links: https:\/\/curtjaimungal.substack.com\n\n\nGuests do not pay to appear. #science","thumbnailUrl":["https:\/\/i.ytimg.com\/vi\/TgFmA-Qwsek\/default.jpg","https:\/\/i.ytimg.com\/vi\/TgFmA-Qwsek\/mqdefault.jpg","https:\/\/i.ytimg.com\/vi\/TgFmA-Qwsek\/hqdefault.jpg","https:\/\/i.ytimg.com\/vi\/TgFmA-Qwsek\/sddefault.jpg","https:\/\/i.ytimg.com\/vi\/TgFmA-Qwsek\/maxresdefault.jpg"],"uploadDate":"2026-05-27T16:05:24+00:00","duration":"PT1H58M18S","embedUrl":"https:\/\/www.youtube.com\/embed\/TgFmA-Qwsek","publisher":{"@type":"Organization","@id":"https:\/\/www.youtube.com\/channel\/UCdWIQh9DGG6uhJk8eyIFl1w#Organization","url":"https:\/\/www.youtube.com\/channel\/UCdWIQh9DGG6uhJk8eyIFl1w","name":"Curt Jaimungal","description":"Exploring theoretical physics, consciousness, philosophy, AI, and God in a technical manner. Visit https:\/\/curtjaimungal.substack.com for early access videos and extra writings.\n\nMy name's Curt Jaimungal, a Torontonian with a degree in mathematical physics from the University of Toronto and I analyze various Theories of Everything from this analytic perspective. The separating factor of TOE from other podcasts is its focus on depth even at the risk of limiting the audience due to how much detail we delve into subjects, paralleling the intensity found in academic discourse. Theories of Everything is known for being the most technical podcast in the physics, math, and philosophy space, as its audience is geared toward academics \/ researchers \/ faculty.\n\nContact toe [at] indiefilmTO [dot] com for business inquiries \/ sponsorship.\n\nGuests do not pay to appear.\n","logo":{"url":"https:\/\/yt3.ggpht.com\/a0bukHxNxRO6puLOjbXzuwfsFjRQ1jHUzn08IzEq4HMYGHQBZG7RDnNWNGZX3u4MNsZcE-K3BQ=s800-c-k-c0x00ffffff-no-rj","width":800,"height":800,"@type":"ImageObject","@id":"https:\/\/www.youtube.com\/watch?v=TgFmA-Qwsek#VideoObject_publisher_logo_ImageObject"}},"potentialAction":{"@type":"SeekToAction","@id":"https:\/\/www.youtube.com\/watch?v=TgFmA-Qwsek#VideoObject_potentialAction","target":"https:\/\/www.youtube.com\/watch?v=TgFmA-Qwsek&t={seek_to_second_number}","startOffset-input":"required name=seek_to_second_number"},"interactionStatistic":[[{"@type":"InteractionCounter","@id":"https:\/\/www.youtube.com\/watch?v=TgFmA-Qwsek#VideoObject_interactionStatistic_WatchAction","interactionType":{"@type":"WatchAction"},"userInteractionCount":86714}],{"@type":"InteractionCounter","@id":"https:\/\/www.youtube.com\/watch?v=TgFmA-Qwsek#VideoObject_interactionStatistic_LikeAction","interactionType":{"@type":"LikeAction"},"userInteractionCount":2182}]},"about":["Insights","\u300eEnglish\u300f"],"wordCount":1723,"keywords":["permanent"]},{"@context":"https:\/\/schema.org\/","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Insights","item":"https:\/\/blog.terabox.com\/insights\/#breadcrumbitem"},{"@type":"ListItem","position":2,"name":"Roman Yampolskiy: AI Safety and the Simulation Hypothesis","item":"https:\/\/blog.terabox.com\/insights\/roman-yampolskiy-ai-safety-simulation-hypothesis#breadcrumbitem"}]}]