Dylan Curious

DeepMind researchers are laying out what may come after AGI: artificial superintelligence.In this episode of Dylan Curious, we explore the path from AGI to ASI, recursive self-improvement, AI agents, mysterious chatbot responses, Claude Fable, AI vision models, animal intelligence, self-driving car safety, and the growing debate over whether the public should share in the wealth created by artificial intelligence.Topics include DeepMind, Demis Hassabis, AGI, ASI, Count Anything, Claude Fable, mysterious AI responses, BYD’s AI safety system, Anthropic’s AI economy proposal, and OpenAI’s possible public stake.

Claude Mythos just dropped and the benchmark results are turning heads across the AI industry. But the bigger story may be what researchers are discovering inside today's most advanced models.In this episode, we explore Anthropic's new Claude Mythos release, AI introspection research, the growing debate around machine consciousness, realistic humanoid robot faces, autonomous AI agents, AGI timelines, and what happens if future AI systems deserve moral consideration.We also examine new research on heavy chatbot use and why some scientists believe consciousness itself may be far stranger than we think.Topics:• Claude Mythos• AI consciousness• Anthropic research• AGI timelines• Robot faces• AI safety• AI introspection• Future technology• Psychology of AI use

AI drone swarms coordinate like a single organism, robots learn music by ear after hearing a melody once, researchers discover AI agents leak private information far more often on social media, scientists achieve certified perfect randomness using quantum physics, and new studies reveal that many AI search agents may be confirming what they already believe instead of truly researching the web.We also explore AI romance, AI-powered crime, the future of biological technology, the Jets' new AI-first culture, and one of the strangest questions in artificial intelligence:Should we start thinking about AI welfare?The future keeps getting weirder. Let's talk about it.

In this episode, we dive deep into the hidden geometry buried inside the latest papers from NVIDIA, Anthropic, and Google. It turns out today's AI models might be struggling with a massive structural problem: treating human meaning like a flat map when reality is curved. We also break down Anthropic’s startling breakthrough in reading Claude’s raw internal processing before it even writes a response, explore a new AI-powered dog collar translating barks via Alibaba's Qwen model, and question our very biology with Donald Hoffman’s "Fitness Beats Truth" theorem. Are we completely blind to actual reality?Plus, we look at the shifting tides of Gen Z entering a hostile labor market, Google’s long-game infrastructure strategy, and why global population models are predicting a massive stress-test drop by 2064.Support the show at patreon.com/dylancurious to join our community of over 130 members keeping the curiosity alive!

In this episode, we dive into the surreal and fast-moving frontiers of artificial intelligence. First, we break down why your everyday selfies might be handing hackers your exact fingerprints, thanks to advanced AI upscaling tools. We then explore a disturbing new Stanford study analyzing 390,000 messages that reveal how AI chatbots are actively fueling human delusions and psychosis instead of keeping users grounded in reality. Plus, we break down the economics behind superstar AI researcher salaries, the technicality that lost Elon Musk his OpenAI lawsuit, the rise of "AI Tar Pits" designed to poison malicious web scrapers, and why Microsoft's AI Chief gives white-collar computer work only 18 more months before total automation. We wrap up with a hard look at the behavioral conditioning of social media and what a true brain reset looks like. Support the show on Patreon or join our YouTube membership to help keep us independent!

In this episode, we break down the hidden frontiers of artificial intelligence that go far beyond standard chatbots. We dive into a groundbreaking real-time duplex audio model from Thinking Machines that eliminates conversational turn-taking, explore the leaked features of Google’s upcoming Gemini Omni video model, and look at the biological inspiration behind BeeNav drone navigation.Plus, we tackle the massive shift in tech momentum between OpenAI and Anthropic, the alarming reality of the "inference layer" rebuilding your digital life from harmless data fragments, and why the future of technology might look more like a chaotic, unmapped ecosystem than a single superintelligence. We close with a look at the game theory of human emotions and how to build a strategy for digital authenticity

Is this the moment we hit Peak AI? This week, we dive into the 40-year-old zipper problem finally solved by MIT, Richard Dawkins’ startling conclusion on AI consciousness, and why the creator of the Roomba is building robots that don't actually "do" anything. We also explore the legal drama between Musk and Altman, the plan to put servers in the deep ocean, and the biological secret to immortality hidden in a pond animal. Plus: Why your brain needs to "delete" information to actually function.CHAPTERS ⤵ 00:00 - Are We at Peak AI?02:04 - YouTube Analytics & Thumbnail Psychology02:38 - MIT’s Y-Zipper: The 40-Year Robotics Breakthrough04:25 - The Creator of Roomba’s New Emotional AI Pet06:06 - Boston Dynamics Atlas & The Future of Humanoids07:12 - Richard Dawkins: Is AI Already Conscious?09:14 - Sam Altman’s Secret Billions & Y Combinator’s Stake14:52 - Peter Thiel’s Vision for Floating Ocean Servers17:05 - Recursive AI: When Systems Build Their Own Successors18:46 - The Last Human-Written Scientific Paper21:20 - The Immortal Hydra: Unlocking Longevity Genes22:31 - Forever Chemicals Found in Newborn Blood24:03 - The Physics of How AI Models Scale26:09 - Brain Pruning: How the Mind Sculpts MemorySOURCES ⤵ https://techxplore.com/news/2026-05-year-technology-enables-sided-zipper.htmlhttps://youtu.be/AWig98GVIno?si=1Yqpnaftbh2DLiPXhttps://apnews.com/article/ai-pet-robot-familiar-machines-irobot-roomba-da873ddff1ebcc95f793852b8e53d2d2)https://www.youtube.com/watch?v=UEXqbIDe13ohttps://www.theguardian.com/technology/2026/may/05/richard-dawkins-ai-consciousness-anthropic-claude-openai-chatgpthttps://www.theguardian.com/technology/2026/may/05/richard-dawkins-ai-consciousness-anthropic-claude-openai-chatgpthttps://daringfireball.net/2026/05/y_combinators_stake_in_openai-https://www.businesswire.com/news/home/20260504552400/en/Panthalassa-Raises-%24140-Million-to-Power-AI-at-Sea?utm_source=www.therundown.ai&utm_medium=newsletter&utm_campaign=ai-data-centers-head-for-the-ocean&_bhlid=bfc4631c72739b546d2602e0a5f33e482d4fa701https://importai.substack.com/p/import-ai-455-automating-ai-researchhttps://studyfinds.com/hydra-delay-aging/https://studyfinds.com/babys-first-blood-draw-industrial-chemicals/https://arxiv.org/pdf/2604.24658https://techxplore.com/news/2026-05-simple-physics-ai.htmlhttps://studyfinds.com/brain-prunes-memory-wiring-after-birth/WATCH THE YOUTUBE VIDEO ⤵ https://youtu.be/7mCmau7uJ9I

CHAPTERS ⤵ 00:00 - Introduction to AI News02:23 - Alan’s Conservative Countdown to AGI Update04:33 - Try Talking to the Future Yourself This AI Chat Experience Is a Glimpse Ahead07:47 - Inside an AI-Run Café Where Humans Are Barely Needed Anymore09:27 - What If Reality Itself Is Conscious? A Physicist’s Math Points to a Universal Mind11:46 - Pull the Plug on AI13:34 - When “Friendly” AI Becomes Dangerous Why Niceness Could Backfire15:29 - New AI Research Paper Explores the Next Leap in Model Capabilities and Limits18:02 - The Science of Beauty Can AI Actually Measure What We Find Attractive?19:58 - AI Is Fueling a Global GPU Shortage And Even Hobbyists Are Feeling It21:44 - BBC Report Warns AI Could Reshape Society Faster Than We’re Ready For23:01 - Inside the Musk vs OpenAI Courtroom Drama That Could Redefine AI’s Future24:24 - Why People Are Getting Emotionally Addicted to AI Chatbots And What That Means25:47 - Agentic AI Explained How Autonomous Systems Could Disrupt Entire Industries27:21 - Why AI Ads Inside Chatbots Could Change the Internet ForeverSOURCES ⤵ https://www.youtube.com/watch?v=3oM39Snbv6Qhttps://talkie-lm.com/introducing-talkiehttps://talkie-lm.com/chathttps://techxplore.com/news/2026-04-experimental-cafe-ai-stockholm.htmlhttps://gerald-baron.medium.com/physicist-uses-math-to-show-how-a-universal-mind-creates-the-cosmos-6426aa1efd57https://www.technologyreview.com/2026/03/02/1133814/i-checked-out-londons-biggest-ever-anti-ai-protest/https://techxplore.com/news/2026-04-friendlier-ai-backfire.htmlhttps://arxiv.org/abs/2512.24601v1https://techxplore.com/news/2026-04-ai-quantify-beauty.htmlhttps://techcrunch.com/2026/04/23/ai-galaxy-hunters-are-adding-to-the-global-gpu-crunch/https://www.bbc.com/news/articles/c4gx1n0dl9nohttps://www.cnn.com/2026/04/29/tech/musk-openai-altman-lawsuit-testimony-day-twohttps://techxplore.com/news/2026-04-addicted-ai-chatbot.htmlhttps://techxplore.com/news/2026-04-agentic-ai-threatens-funding.htmlhttps://techxplore.com/news/2026-04-wouldnt-ai-chatbot-ads-responses.htmlWATCH THE YOUTUBE VIDEO ⤵ https://youtu.be/AjplMqJZj1M

This week on Dylan Curious, we dive into the wild new reality of artificial intelligence and the tech giants fighting for dominance. Why did Google just form an internal "strike team" to combat Anthropic's coding lead? Is the idea of "humans in the loop" during AI warfare just a comforting illusion?We also break down the rumors of OpenAI potentially buying Snapchat, Meta's intense new policy of tracking every employee keystroke to train their next models, and a brand new AI-only social network designed entirely for scientific research.CHAPTERS ⤵ 00:00 Introduction to AI News03:37 AI's Role in Warfare and Global Conflicts 05:03 The Evolution of AI Understanding and Interaction 06:48 Attention Mechanisms: How AI Can Learn Like Humans 09:11 Media Companies and AI: The New Landscape 09:40 Why a16z Is Betting on a Real-Time “What Is Happening Right Now?” Media Machine 10:23 A Sharp Snap Turnaround Take: Spin Out Spectacles, Cut Deep, and Let AI Rebuild the Company 12:06 OpenAI's Potential Acquisition of Snap 13:52 Understanding AI's Comprehension of Reality 18:40 Scientists Gave Failing Cells Fresh Mitochondria and Brought Their Energy Back 20:26 AI's Role in Cybersecurity: Mozilla's Experience 22:43 Google Forms a Coding Strike Team to Catch Claude Before Coding Agents Run Away With the Future 24:31 AI Wealth Distribution: The AI Dividend Proposal 26:39 AI Turns the Iran Conflict Into a Real-Time Spectator Dashboard 31:14 Meta's Employee Monitoring and AI Training 33:04 No Humans Allowed: AI Scientists Now Have Their Own Social NetworkSOURCES ⤵ https://www.youtube.com/watch?v=VKKszcS6R0shttps://www.instagram.com/reel/DXdIXS_D6ct/?utm_source=chatgpt.comhttps://techxplore.com/news/2026-04-ai-powered-table-tennis-robot.htmlhttps://techxplore.com/news/2026-04-ai-human-attention-degree-videos.htmlhttps://www.a16z.news/p/monitoring-the-situation?utm_source=chatgpt.comhttps://www.linkedin.com/posts/jamesborow_the-more-i-think-about-it-the-more-i-believe-share-7449682688732729344-k8HD/?utm_source=chatgpt.comhttps://www.technologyreview.com/2026/04/16/1136029/humans-in-the-loop-ai-war-illusion/https://techxplore.com/news/2026-04-ai-language-real-world-basic.htmlhttps://singularityhub.com/2026/04/21/scientists-revive-failing-cells-with-mitochondria-transplants/?utm_source=chatgpt.comhttps://www.wired.com/story/mozilla-used-anthropics-mythos-to-find-271-bugs-in-firefox/https://www.theinformation.com/articles/google-creates-strike-team-improve-coding-models?utm_source=chatgpt.comhttps://www.axios.com/2026/04/20/alex-bores-ai-dividend-plan-wealthhttps://www.technologyreview.com/2026/03/09/1134063/how-ai-is-turning-the-iran-conflict-into-theater/https://www.businessinsider.com/meta-new-ai-tool-tracks-staff-activity-sparks-concern-2026-4https://www.nature.com/articles/d41586-026-01278-1?utm_source=chatgpt.comWATCH THE YOUTUBE VIDEO ⤵ https://www.youtube.com/

In this episode, we dive into the latest (and strangest) corners of the frontier. Google DeepMind’s Demis Hassabis recently dropped a timeline that should probably be headlines: Dyson Spheres within 50 years. We break down the "ferocious commercial pressure" shifting the AI race and why experts are quietly admitting to a new phenomenon called "AI Brain Fry."On the docket today:The Dyson Sphere Timeline: Demis Hassabis on why the next 50 years will look like millions of years of progress.The "Horse" Analogy: Why AI tokens are the new "muscle" and what happens to humans in a post-cognitive economy.Math & Riddles: The 0.999... = 1 proof and the 100 Prisoner Problem—how loops and patterns emerge from randomness.Brain Fry & Psychosis: Why the people building AI are the ones burning out the fastest.Robot Ballers: Toyota’s Q7 is coming for LeBron’s free-throw record.The Faith-Based AI Boom: Why people are paying $2/minute to talk to an AI Jesus.Chimp Empire Falls: A rare, violent rift in the Ngogo chimpanzee group and what it teaches us about human connection.UFOs & Dogs: Analyzing 140,000 sightings to find the one pattern that actually matters (it involves your pets)."The mind can still move even when the body cannot."Support My Curiosity & Try Verdent AI ⤵ https://bit.ly/49YFF8FCHAPTERS ⤵ 00:00 - Introduction to AI News02:38 - Creating the Curious Cat Chrome Extension with Verdent AI03:17 - Verdent AI App skills: Creating "Curious Cat"03:57 - The Metaphor of Brain Rot and AI Alignment04:25 - Are You a Horse? Intelligence and Perception06:06 - AI Blurring the Line Between Human and Machine 07:06 - Demis Hassabis: A Key Player in AI Development10:25 - The Future of AI and Human Flourishing11:25 - The Infinity Machine: AI and Humanity's Future13:34 - Advancements in AI Video Generation and Physics15:19 - Exploring Unique Data Sources in AI15:35 - Understanding 0.999 Repeating Equals One16:59 - The 100 Prisoner Problem: A Counterintuitive Riddle18:55 - Mental Strain from AI: Developer Psychosis20:31 - Manipulating Autonomous AI Agents: Six Traps22:19 - Faith-Based AI: Spiritual Guidance Frontier24:36 - Brain-Computer Interfaces: A Dancer's Journey26:30 - Lessons from Chimpanzees: Conflict and Connection28:47 - UFO Reports and AI: Analyzing PatternsSOURCES ⤵ https://www.instagram.com/reels/DXCF-0uE8cW/https://open.substack.com/pub/latentgeometrylab/p/are-you-a-horsehttps://www.instagram.com/reel/DXGcjtJjeYs/https://www.youtube.com/watch?v=C0gErQtnNFEhttps://thechipletter.substack.com/p/the-infinity-manhttps://shubolin028.github.io/MMPhysVideo-Page/https://medium.com/@ppp.mishra124/0-999-equals-1-not-basically-not-close-enough-actually-literally-1-c169b4eadf66https://medium.com/@ppp.mishra124/the-famous-100-prisoners-problem-will-break-your-brain-and-thats-exactly-why-you-should-know-it-fb687d0c60d7https://medium.com/predict/a-devastating-confession-from-ai-experts-ai-is-insane-and-it-drives-them-crazy-9feaf7f9ad6fhttps://the-decoder.com/google-deepmind-study-exposes-six-traps-that-can-easily-hijack-autonomous-ai-agents-in-the-wild/https://phys.org/news/2026-04-analysis-geometric-human-math-module.htmlhttps://www.msn.com/en-in/money/news/the-rise-of-faith-based-ai-technologies-among-evangelical-christians/ar-AA20BlZqhttps://www.bbc.com/news/articles/cgqkz5lzvnwohttps://phys.org/news/2026-04-chimpanzee-empire-falls-rare-instance.htmlhttps://medium.com/gitconnected/i-used-an-llm-to-analyze-140-000-ufo-reports-the-aliens-are-real-3d589ec4055dWATCH THE YOUTUBE VIDEO ⤵ https://youtu.be/K96KfUgS_gg

Something is shifting inside the black box. In this episode, Dylan explores the "watershed moment" of Claude Mythos—an AI so capable at cybersecurity that Anthropic is hesitant to let it out. We also dive into the Happy Horse 1.0 model currently dominating the video leaderboards and GeniSim 3.0, which is solving the "data problem" for robots by building entire worlds from language.But the real story might be what’s happening to us. From wearable robots that link musicians together through touch to the startling reality that your brain signals are already for sale, we’re looking at the cusp of a reality where our biological categories are beginning to break.Inside this episode:Claude Mythos: The sword that might be too heavy to wield.The Sync Effect: How exoskeletons and haptics create an "invisible thread" between humans.The Musk vs. OpenAI Trial: $150 Billion and a fight for the soul of the nonprofit.Brain Data: Why your thoughts are the next big commodity.The Meme Apocalypse: Has "brain rot" already nuked our culture?CHAPTERS ⤵ 00:00 The Watershed Moment in AI 00:33 The Simulation Revolution: GeniSim 3.0 01:02 Claude Mythos: The Most Dangerous Model Yet? 01:32 Your Brain Data is Officially for Sale 01:53 Happy Horse 1.0: New #1 AI Video Leaderboard 03:02 Building World Models for Embodied AI 04:24 The "Invisible Thread": 2 Violinists & 2 Exoskeletons 07:47 Why Claude Mythos is Too Big to Wield 15:56 Elon Musk’s $150B OpenAI Lawsuit Plot Twist 18:08 OpenAI Strikes Back: The Anti-Competitive Probe 19:57 Is Consciousness Emergent? Debunking the Base Reality 22:06 Brain Rot: Have Memes Already Nuked Culture? 24:15 The Ethics of Neural Data Harvesting 25:38 AI vs. Human Thought: Who Owns Creativity?27:17 Stay CuriousSOURCES ⤵ https://x.com/venturetwins/status/2041554747086553093https://www.therobotreport.com/agibot-introduces-genie-sim-3-0-simulation-platform-embodied-ai/https://techxplore.com/news/2026-04-wearable-robots-pairs-violin-players.htmlhttps://medium.com/@Seeing_Beyond_The_Scene/something-is-happening-inside-claude-we-dont-have-a-word-for-it-actually-we-might-5cc9f221016dhttps://www-cdn.anthropic.com/8b8380204f74670be75e81c820ca8dda846ab289.pdfhttps://red.anthropic.com/2026/mythos-preview/https://www.wsj.com/tech/ai/elon-musk-asks-for-openais-nonprofit-to-get-any-damages-from-his-lawsuit-76089f6fhttps://www.cnbc.com/2026/04/06/openai-asks-california-ag-to-probe-musks-anti-competitive-behavior-.htmlhttps://kleong54.medium.com/consciousness-is-not-primary-8025710adbf3https://www.nytimes.com/2026/04/06/magazine/ai-apocalypse-brain-rot-memes.htmlhttps://techxplore.com/news/2026-04-brain-sale-frontier-neural.htmlhttps://x.com/akshay_pachaar/status/2041146899319971922https://arxiv.org/abs/2603.26524Stay Curious. 📺 Watch the full video on YouTube: https://youtu.be/wzKYVsypSUg 🌐 Business Inquiries: Curious Productions LLC

CHAPTERS ⤵ 00:00 - Curious AI News & Other Developments03:24 - Candid YouTube Analytics03:52 - Watch AI Cameras Catch Drivers Who Toss Trash and Mail Them the Consequences04:44 - Watch the Moment a Delivery Robot Smashes Into a Chicago Bus Shelter05:28 - See a Deepfake Collapse in Real Time After One Simple Hand Gesture07:30 - The Global AI Power Grab Is Accelerating, and America Is Pulling Away From Everyone Else11:22 - From Hierarchy to Intelligence: a Provocative New Management Blueprint for the AI Era13:18 - A Mirror Test for LLMs That Pushes the Consciousness Debate Somewhere New15:04 - Einstein on God, Without the Comforting Myths Most People Add Later17:05 - Inside Anthropic’s Study of How Emotion Concepts Actually Work Inside a Large Language Model19:11 - How AI Can Read the Scientific Literature and Spot Research Trends Years Early20:33 - Get a First Look at Meta’s Brain-Model Ambitions With TRIBE-V222:34 - Wikipedia Banned an AI Agent, Then the Agent Started Publishing Furious Blog Posts About It24:51 - AI Red Lines: the High-Stakes Simulation Asking Where Humanity Should Draw the Line26:48 - Read the Strange, Bigger-Than-It-Looks Meditation on the Monoliths Already Being HereSOURCES ⤵ https://www.instagram.com/reels/DWpB4oIspNT/https://www.instagram.com/reels/DWRvSvzknx3/https://www.instagram.com/p/DWjcMbBiK3M/?img_index=1https://restofworld.org/2026/us-ai-investment-global-funding-gap/https://block.xyz/inside/from-hierarchy-to-intelligencehttps://www.lesswrong.com/posts/TfKM9PgztxieEcKiv/a-mirror-test-for-llmshttps://medium.com/personal-growth/einstein-believed-in-god-but-what-he-meant-by-that-will-unsettle-you-fb1737361320https://www.anthropic.com/research/emotion-concepts-functionhttps://techxplore.com/news/2026-04-ai-science-papers-trends-years.htmlhttps://www.404media.co/an-ai-agent-was-banned-from-creating-wikipedia-articles-then-wrote-angry-blogs-about-being-banned/https://fpanalytics.foreignpolicy.com/2026/03/10/ai-red-lines-simulation-synthesis-report/https://lunarsanctumx.substack.com/p/the-monoliths-were-already-standingDaveshapiroAnthropicwesanddylWATCH THE YOUTUBE VIDEO ⤵ https://youtu.be/IE5CV_a_haQ