AI Convo Cast

In this episode, we discuss OpenAI's restricted preview of the GPT 5.6 family including Sol, Terra, and Luna, and why the rollout is gated to about twenty government-approved companies. We cover Anthropic getting a partial green light from the U.S. Commerce Department to bring Claude Mythos 5 back for trusted users, and what this new "approved list" era means for frontier AI access. We also dig in...

In this episode, we discuss OpenAI's new Codex usage data showing AI agents taking on multi-hour delegated work, Google baking computer use directly into Gemini 3.5 Flash, and Figma's Config 2026 reveal of code layers that fuse design and production code. We also break down Qualcomm's aggressive AI data center push with Meta and Microsoft as launch customers, the new Dragonfly C1000 inference chip...

In this episode, we discuss OpenAI's first custom inference chip Jalapeño built with Broadcom, and what it means for OpenAI's push to cut compute costs and reduce reliance on Nvidia. We dig into Anthropic's accusations that operators tied to Alibaba and the Qwen AI lab ran the largest known Claude distillation attack, the Trump administration pressing Meta to join voluntary pre-release AI model re...

In this episode, we discuss Anthropic's new Claude Tag agent for Slack, Meta's $299 displayless AI Glasses powered by Muse Spark, Google DeepMind's filmmaking partnership and $75 million investment in A24, and Reflection AI's $6.3 billion compute deal with SpaceX's xAI Colossus 2 facility. We break down how Claude Tag introduces persistent, multiplayer AI memory inside enterprise Slack, why Meta i...

In this episode, we discuss OpenAI's Daybreak cybersecurity expansion with GPT 5.5 Cyber and the Patch the Planet program, Nvidia's bold London Climate Week claim that warm liquid cooling can solve AI's data center water problem, and Google's Gemini API crackdown that's breaking apps using unrestricted standard keys. We also dig into Anthropic's controversial privacy policy update, which could req...

In this episode, we discuss OpenAI's major push into real-world science with LifeSciBench and a GPT 5.4 chemistry breakthrough that directed a wet lab to improve a notoriously tricky medicinal chemistry reaction. We also break down Google DeepMind's new AI Control Roadmap that treats powerful AI agents like potential insider threats, Google's abrupt decision to kill the open-source Gemini CLI in f...

In this episode, we cover the G7's new "trusted partners" framework for advanced AI access and Anthropic's signal that blocked frontier models could return to allied users soon. We break down a striking DOJ filing framing xAI's Grok as critical military infrastructure inside the Pentagon's Project Maven, Google's new Gemini-powered Home Speaker launch, and Amazon's major agent push at AWS Summit N...

In this episode, we discuss the G7's emergency talks over U.S. frontier AI access after the Trump administration ordered Anthropic to disable its Fable 5 and Mythos 5 models, Microsoft making Copilot Cowork generally available with new usage-based Copilot Credits pricing, and xAI shipping Grok Imagine Video 1.5 with synchronized audio to compete with Google's Veo and OpenAI's Sora. We also break d...

In this episode, we discuss Anthropic's standoff with the U.S. Commerce Department over its Fable 5 and Mythos 5 models, which were pulled offline globally after a disputed jailbreak claim. We also break down Microsoft's Work IQ APIs hitting general availability, turning Microsoft 365 into a substrate for third-party agents with MCP-style tools and Copilot Credits billing. Next, we cover Google's...

In this episode, we discuss the cybersecurity community's open letter pushing back on the U.S. export ban of Anthropic's Fable and Mythos models, Meta launching AI Mode on Facebook search powered by its Muse Spark model and public posts, and AWS rolling out FinOps Agent, adding Google's Gemma 4 to Bedrock, and claiming major productivity gains from agentic "frontier teams." We also cover Loft Orbi...

In this episode, we discuss Anthropic's Claude Fable 5 and Mythos 5 being yanked offline after a Trump administration export directive, and the company rushing technical staff to Washington to negotiate access back. We also break down a multistate attorneys general investigation into OpenAI covering sycophancy, engagement design, and protections for minors as the company eyes an IPO. Finally, we c...

In this episode, we discuss Google's new DiffusionGemma model that generates text in parallel blocks for up to 4x faster inference, Anthropic walking back silent model downgrades in Fable 5 after developer backlash, OpenAI reportedly planning aggressive API price cuts to pull developers away from Anthropic, and a 12-hour Gemini outage that exposed how deeply AI assistants have embedded into produc...

In this episode, we discuss Anthropic's new Claude Fable 5 model drawing backlash from cybersecurity researchers over aggressive safety filters, Google slashing the price of its AI Plus subscription with Gemini and NotebookLM bundled in to ignite an AI price war, and Decart launching Oasis 3 as a programmable world model API for autonomous vehicles and robotics. We also break down a new lawsuit fr...

In this episode, we discuss Anthropic's public release of Claude Fable 5 from its previously restricted Mythos class, Apple's long-awaited Siri AI reveal at WWDC 2026, and a malware attack on Microsoft's open source GitHub repos that targets AI coding workflows in tools like Claude Code, Gemini CLI, and VS Code. We also dig into Lovable crossing $500 million in annualized revenue as nontechnical f...

In this episode, we cover OpenAI's confidential S-1 filing and its "third phase" AGI roadmap, Apple's long-awaited Siri AI overhaul powered in part by Google Gemini, Google's massive $920 million per month GPU bridge deal with SpaceX, and Notion temporarily cutting off Anthropic's Claude models after an Opus outage. We unpack what OpenAI going public could reveal about its finances and safety risk...

In this episode, we discuss OpenAI's reported plan to transform ChatGPT into an AI super app with integrated Codex, agents, and partner services like Canva and Booking. We break down Anthropic's striking claim that Claude now writes roughly 80 percent of its own code, along with its call for a coordinated industry pause mechanism on frontier AI development. We also cover a rare coalition of AI lea...

In this episode, we discuss OpenAI's upgraded GPT Rosalind model for life sciences research, the new Dreaming V3 memory architecture rolling out to ChatGPT users, and Google DeepMind's release of Gemma 4 12B, an open-weight multimodal model built to run locally on laptops. We also dig into a striking report that Apple's revamped Siri will reportedly lean on Google Gemini running on NVIDIA Blackwel...

In this episode, we cover Meta's global launch of its Business Agent across WhatsApp, Messenger, and Instagram, and OpenAI's pivot of Codex from a coding tool into a general productivity platform now serving millions of weekly users. We also break down a troubling security incident where Instagram's AI support chatbot was reportedly tricked into helping attackers take over accounts, and the UK's l...

In this episode, we cover Microsoft's Build 2026 launch of Scout and its new MAI family of in-house models, including the MAI-Thinking-1 reasoning model trained with "zero distillation." We also break down OpenAI turning Codex into a workplace agent with role-specific plugins for analytics, sales, design, and finance, plus the general availability of GPT-5.5, GPT-5.4, and Codex on Amazon Bedrock....

In this episode, we cover NVIDIA's launch of Cosmos 3, an open foundation model for physical AI spanning robotics, autonomous vehicles, and simulation. We break down the NVIDIA and Microsoft partnership behind RTX Spark and the Surface Laptop Ultra, which can run 120 billion parameter models locally and reframes the PC as an agent host. We also dig into MiniMax's surprise M3 release, an open-weigh...

In this episode, we discuss OpenAI bringing Codex Computer Use to Windows, letting the agent see, click, and type through Windows apps while developers steer it remotely from ChatGPT on mobile or Codex on Mac. We break down the developer backlash to GitHub Copilot's shift to token and AI-credit-based billing, the new Pro, Pro Plus, and Max tiers, and why agentic workflows are breaking the flat-sub...

In this episode, we discuss Anthropic's launch of Claude Opus 4.8 with new Dynamic Workflows in Claude Code, enabling hundreds of parallel subagents for long-running agentic tasks. We break down Anthropic's massive $65 billion Series H funding round at a near-trillion dollar valuation, the enormous compute deals with Amazon, Google, Broadcom, and SpaceX, and what it signals about the next phase of...

In this episode, we discuss Robinhood launching Agentic Trading and an Agentic Credit Card that let AI agents place real stock trades and swipe virtual cards through MCP-compatible tools like Claude, Cursor, and Codex. We also cover YouTube's new automatic AI labeling system for photorealistic generative content, including videos made with Google's Veo, and how it shifts AI provenance from an hono...

In this episode, we cover Anthropic's candid engineering post on containing Claude agents and capping the "blast radius" of autonomous AI, plus Demis Hassabis tightening DeepMind's AGI timeline to as early as 2029 and calling today's agent era a "practice run." We also dig into the Gemini API key security mess, including a $10,000 surprise bill and Google Cloud's warnings about shadow AI, and revi...

In this episode, we discuss Google DeepMind's AlphaProof Nexus autonomously resolving long-standing Erdős problems using the Lean proof assistant, marking a real shift from olympiad puzzles to research-level mathematics. We break down Anthropic's Project Glasswing, where Claude helped partners like Cloudflare and Mozilla surface over 10,000 high-severity software vulnerabilities, flipping the trad...