Robots That Learn Jobs by Watching You

AI is finding software vulnerabilities faster than organizations can patch them. OpenAI reported 948 Linux vulnerabilities with working privilege-escalation exploits. Should we be worriedJoin The Neuron's Corey Noles LIVE with Mark Stockley, Cybersecurity Evangelist at ThreatDown and host of The AI Fix podcast, to explore what AI-powered vulnerability discovery means for businesses, developers, and everyday users.Thanks to Alumni Ventures for sponsoring this episode and helping accredited individual investors access professionally managed venture capital portfolios. Explore their investment opportunities and learn more at https://www.av.vc/.From OpenAI's Linux research to PHP2Shell and Anthropic's Project Glasswing, we'll separate the real risks from the hype.What We'll Cover:- OpenAI's 948 findings: What was actually demonstrated?- PHP2Shell: Why PHP flaws can threaten web servers.- The Patch Tsunami: Can defenders keep up with AI discovery?- Claude Mythos: Why access to some AI models is restricted.- AI cybercrime: Could attackers exploit the patching gap?- CVE counts: Does more disclosure mean less security?- Next steps for users, developers, and security teams.Guest & Resources:- ThreatDown: https://www.threatdown.com/- The AI Fix: https://theaifix.show/- The Patch Tsunami: https://www.threatdown.com/blog/the-patch-tsunami/- OpenAI research: https://openai.com/index/a-thousand-fewer-paths-to-root/- PHP updates: https://www.php.net/- Project Glasswing: https://www.anthropic.com/project/glasswing🔔 Subscribe for clear, practical AI coverageAI-powered vulnerability discovery is outpacing our ability to patch systems. Learn if your business is at risk today.Mark Stockley joins Corey Noles to examine the real-world implications of automated hacking. We analyze whether the current patch tsunami is a manageable challenge or an existential threat to modern web security. The discussion covers why OpenAI research and other automated tools are finding flaws at an unprecedented scale, and what developers must prioritize to maintain system integrity.We also separate genuine cybercrime risks from marketing noise, including the technical reality behind PHP vulnerabilities and restricted AI model access. By understanding these patterns, security teams can better defend against potential exploits.Subscribe for practical AI coverage and let us know in the comments how your team manages critical software updates.

Corey and Grant walk backward through roughly four years of modern generative AI, starting with late-2026 frontier models and agents and tracing the path back to ChatGPT’s November 2022 launch and the 2017 Transformer paper. The episode focuses on what changed at each stage: reasoning, tools, multimodality, long context, agents, open-source competition, and the shift from chatbots toward long-running computer work. They also revisit their own first encounters with ChatGPT, debate open access versus AI safety, and close with long-term predictions about robotics, machine labor, scientific progress, and who should have access to increasingly capable systems.Subscribe to The Neuron newsletter: https://theneuron.aiSponsored by ZoomInfo. Learn more: https://www.zoominfo.com/theneuron

Corey talks with Napster CEO John Acunto about the company’s pivot from music into AI agents and what Napster calls “streaming intelligence.” Acunto lays out a four-part customer-data framework, capture, learn, act, own, and argues that companies should use conversational agents to learn directly from customers while keeping control of the resulting data and models. The conversation covers photorealistic video agents, edge compute, Microsoft Azure / Foundry, robotics, data portability, customer conversion, and Napster’s broader bet that the next interface for AI will feel more like talking to a person than typing into a box.Learn more: https://napster.comSubscribe to The Neuron newsletter: https://theneuron.aiSponsored by Outshift:“Scaling Out Superintelligence” Vijoy Pandey, January 2026. The technical whitepaper detailing the Internet of Cognition architecture, three-layer stack, and cognition state protocols. https://outshift.cisco.com/internet-of-cognition/whitepaper?utm_campaign=fy27q1_outshift_ww_paid_ioc-neuron-wp_podcast&utm_channel=podcast&utm_source=podcastInternet of Cognition Interactive Demo Clickable walkthrough showing per-agent activity, intent, context, and collective reasoning across a multi-agent SRE system.https://outshift.cisco.com/internet-of-cognition/explore?utm_campaign=fy27q1_outshift_ww_paid_ioc-neuron-explore_podcast&utm_channel=podcast&utm_source=podcast

Standard Bots CEO Evan Beard explains why industrial robots are finally starting to move beyond brittle point-by-point programming. Instead of teaching a robot one exact path, teams can increasingly describe a task, demonstrate it through teleoperation or guided motion, and let the system learn how to handle variation in the real world. The conversation covers natural-language robot programming, learning by demonstration, ugly factory edge cases, the reliability bar for production, and why the most important robotics progress may look less like a humanoid stunt and more like a machine that can handle a shifted part without calling an engineer.Relevant Article Links:https://standardbots.com/Subscribe to The Neuron newsletter: https://theneuron.aiSponsored by Outshift by Cisco. Learn more: https://outshift.comSponsored by Origin. Learn more: https://originhq.com/AIexplained

Neurosurgeon-scientists Alex Ksendzovsky and Jon Pomeraniec explain how The Biological Computing Company uses living neurons as a research platform for AI. Their team grows roughly 150,000 cells on a 4,096-electrode array, sends in patterns such as handwritten digits, records how the neurons respond, and models useful computational behaviors in software. They say those neuron-inspired techniques are already helping GPU-based video-generation systems run about 5x faster with roughly 80% lower inference cost, while their longer-term work explores robots, world models, and real-time biological compute.Relevant Article Links: https://app.tbc.cohttps://www.youtube.com/watch?v=R4eKRihw9f4Subscribe to The Neuron newsletter: https://theneuron.aiSponsored by Mercor: Check out the AI Productivity Index (APEX) here: https://www.mercor.com/apex/?utm_source=neuron&utm_medium=podcast&utm_campaign=apex&utm_content=10_7Sponsored by Outshift:“Scaling Out Superintelligence” Vijoy Pandey, January 2026. The technical whitepaper detailing the Internet of Cognition architecture, three-layer stack, and cognition state protocols. https://outshift.cisco.com/internet-of-cognition/whitepaper?utm_campaign=fy27q1_outshift_ww_paid_ioc-neuron-wp_podcast&utm_channel=podcast&utm_source=podcastInternet of Cognition Interactive Demo Clickable walkthrough showing per-agent activity, intent, context, and collective reasoning across a multi-agent SRE system.https://outshift.cisco.com/internet-of-cognition/explore?utm_campaign=fy27q1_outshift_ww_paid_ioc-neuron-explore_podcast&utm_channel=podcast&utm_source=podcast

OpenAI just dropped 20+ announcements at DevDay 2026, so... yeah, we have some things to talk about. 😺We’re going to break down what OpenAI announced, what actually matters, and what could change how you use AI.🤖 OPENAI DEVDAY 2026OpenAI’s biggest reveal was Dots, always-on AI agents powered by GPT-6 Astra. They get their own cloud computer, connect to 4,000+ apps, learn how you work, and keep working in the background.OpenAI also unveiled ChatGPT Space, collaborative Pages and Slides, Codex in the cloud, the Decisions API, plugin extensions, a $500 Pro plan, and more.DevDay recap:https://openai.com/index/devday-2026-recap/Then there’s GPT-6.1 Sol, which OpenAI says gets close to Astra-level performance on coding, computer use, and professional work at roughly one-fifth of Astra’s standard token price.https://openai.com/index/introducing-gpt-6-1-sol/📰 AND THAT’S NOT ALL...We’ll also get into:⚡ Claude Sonnet 5.5, Anthropic’s faster, cheaper Sonnet update.🌎 AMD buying Fei-Fei Li’s World Labs for $8.2B.🛡️ NVIDIA’s new agent safety system with hardware-level monitoring.🏢 Meta’s new enterprise AI business around Muse, Muse Code, Business Agent, and more.🤖 Manus 2.0 + Cue, including personal agents with their own computer, phone number, email, and wallet.🔥 Plus whatever AI news happens before we go live.What happened? What can you actually use? And are we about to have tiny armies of AI employees running around the internet for us?Come hang, ask questions, and disagree with us in chat. 👇🤖 OpenAI DevDay 2026:https://devday.openai.com/🎥 Official DevDay keynote:https://www.youtube.com/watch?v=Fls_onRviPM😺 Get The Neuron:https://www.theneuron.ai/

Runway co-founder Anastasis Germanidis explains Solaris, an “interface world model” that generates software frame by frame instead of rendering a traditional interface from HTML and CSS. The model reacts directly to clicks, drags, and keyboard input, while language models or agents can still handle logic and tool calls behind the scenes. The conversation explores personalized interfaces, local-versus-cloud world-model architecture, robotics simulation, and Runway’s longer-term bet that world models could become a new foundation for software, games, and physical AI.Relevant Article Links:• Runway Solaris: https://runway.com/news/research/introducing-solaris• Solaris paper: https://arxiv.org/abs/2609.00776• Runway GWM Worlds: https://runway.com/research/introducing-gwm-worlds-2• Robot policy evaluation: https://runway.com/research/accelerating-robot-policy-evaluationSubscribe to The Neuron newsletter: https://theneuron.aiSponsored by Dell Technologies and NVIDIA. Learn more at techrepublic.com/hubs/the-enterprise-guide-to-scalable-ai/.

OpenAI DevDay 2026 is here, and we’re hosting a watch party to watch the keynote LIVE.Join Grant Harvey and Corey Noles from The Neuron as we react to OpenAI’s biggest DevDay announcements in real time, figure out what’s actually new, and translate the demos into what they mean for the rest of us.📅 OPENAI DEVDAY KEYNOTE:Tuesday, September 2910:00 AM PT / 12:00 PM CT / 1:00 PM ETFeaturing Sam AltmanWe’ll be following OpenAI’s stream and weighing in as everything gets announced.New models? Agents? Developer tools? Something completely unexpected?Whatever OpenAI ships, we’ll break it down, explain the demos in normal-person English, and call out what actually matters.What to expect:🤖 Live reactions to OpenAI’s biggest DevDay announcements🧠 Plain-English explanations of what each new model, tool, or API actually does💼 What developers, businesses, and everyday AI users should pay attention to🔍 Our take on what matters versus what just looks good in a demo👀 The inevitable Grant + Corey speculation when OpenAI shows something weirdOfficial OpenAI DevDay page + livestream:https://devday.openai.com/Get our full AI news breakdown in The Neuron:https://theneuron.ai

Corey and Grant sit down with Mammoth Biosciences co-founder and CEO Dr. Trevor Martin to explain how CRISPR is moving from a laboratory tool toward one-time medicines. Trevor breaks down how Mammoth searches massive biological datasets for new CRISPR systems, why AI is useful but still constrained by missing clinical data, and what it would take to build a feedback loop between AI models, wet labs, and human trials. The conversation also covers personalized medicine, the economics of one-time cures, organoids, and why a medical breakthrough may be the clearest way for AI to prove its value to the public.Mammoth Biosciences - https://mammoth.bioSubscribe to The Neuron newsletter: https://theneuron.aiSponsored by Mercor: Check out the AI Productivity Index (APEX) here: https://www.mercor.com/apex/?utm_source=neuron&utm_medium=podcast&utm_campaign=apex&utm_content=10_7Sponsored by Outshift:“Scaling Out Superintelligence” Vijoy Pandey, January 2026. The technical whitepaper detailing the Internet of Cognition architecture, three-layer stack, and cognition state protocols. https://outshift.cisco.com/internet-of-cognition/whitepaper?utm_campaign=fy27q1_outshift_ww_paid_ioc-neuron-wp_podcast&utm_channel=podcast&utm_source=podcastInternet of Cognition Interactive Demo Clickable walkthrough showing per-agent activity, intent, context, and collective reasoning across a multi-agent SRE system.https://outshift.cisco.com/internet-of-cognition/explore?utm_campaign=fy27q1_outshift_ww_paid_ioc-neuron-explore_podcast&utm_channel=podcast&utm_source=podcast

GPT-6 Sol, GPT-6 Luna, AND Claude Opus 5.5 all just dropped. So which model should you actually use?Join Grant Harvey and Corey Noles from The Neuron LIVE today as we put the newest models from OpenAI and Anthropic head-to-head.OpenAI says GPT-6 Sol makes roughly half as many mistakes as GPT-5.6 Sol, bringing Astra-level reliability to a much cheaper model, while GPT-6 Luna is designed for fast, high-volume work like summarization, extraction, and quick questions. OpenAI is also dramatically cutting prices, with Sol reportedly coming in at $2 / $10 per million input / output tokens and Luna at just $0.10 / $0.50.Meanwhile, Anthropic launched Claude Opus 5.5 just 90 minutes earlier. Anthropic says it performs around Claude Fable 5.1 level on most work while costing roughly 40% less to run than Opus 5, generating output more than 30% faster, and communicating more clearly. Its API price is $4 / $20 per million input / output tokens.And the benchmark fight is already spicy: Anthropic reports Opus 5.5 scoring 66.4% on Terminal-Bench 4.0, 54.4% on FrontierCode, and 1,846 Elo on GDPval-AA v2.1, with particularly strong results in coding, computer use, and professional knowledge work.On the livestream, we’ll break down:⚔️ GPT-6 Sol vs. GPT-6 Luna🤖 Claude Opus 5.5 vs. OpenAI’s newest models💻 Coding, reasoning, writing, agents, and everyday work💰 Pricing and what it means for AI users🎯 Which model fits which kind of task🧪 First reactions and live testing💬 Your questions from the chatIf you’ve been wondering whether to use GPT-6 Sol, GPT-6 Luna, Claude Opus 5.5, or GPT-6 Astra, come hang out and test them with us.🔴 LIVE: 3 PM CT / 1 PM PTSubscribe to The Neuron newsletter: https://theneuron.aiModels mentioned:GPT-6 Sol + Luna: https://openai.com/index/introducing-gpt-6-sol-and-luna/Claude Opus 5.5: https://www.anthropic.com/claude-opus-5-5