AI News: Anthropic Leak Shows Us The Future of AI

This podcast delivers a weekly roundup of AI news, separating significant developments from hype. It covers Anthropic's Claude code leak, revealing their work on "Kairos," an autonomous background agent. The host also discusses ReCraft's visually advanced AI model, OpenAI's massive funding round and shift towards a unified "super app," and Microsoft's new transcription model, MAI Transcribe 1, which outperforms competitors. Additional topics include Google's cost-effective video model, AI-driven features in Google Inbox, and new open-source language models from Google, Alibaba, and RC. The episode concludes with rapid-fire updates on Claude, ChatGPT, OpenAI's acquisition of TBPN, Perplexity's tax AI, Slack's AI enhancements, and AI applications in car design and shopping carts.

Outlines

Part 1: Anthropic's Code Leak and Memory Architecture

Part 2: Creative Assets and OpenAI's Strategic Moves

Part 3: Microsoft and Google Model Updates

Part 4: Global Multimodal and Open Source Models

Part 5: Integration, Acquisitions, and Productivity Tools

Part 6: Industrial and Retail AI Applications

Sign in to continue reading, translating and more.

Continue

Matt Wolfe

Part 1: Anthropic's Code Leak and Memory Architecture

Anthropic's Claude Code Leak: Insights into Memory Architecture and Kairos

Kairos: Anthropic's Proactive AI Agent for Autonomous Task Execution

Anthropic's Response to the Leak and Potential Claude Code Spinoffs

Part 2: Creative Assets and OpenAI's Strategic Moves

ReCraft V4: AI Models for Professional-Grade Visual Assets

OpenAI's Massive Raise and Plans for a Unified AI Super App

OpenAI's Agentic Focus, Sora Shutdown, and Microsoft's New Transcription Model

Part 3: Microsoft and Google Model Updates

Testing Microsoft's MAI Transcribe 1 and Availability in Microsoft Foundry

Google's VO 3.1 Lite and AI Inbox for Ultra Subscribers

New Open Source LLMs: Google's Gemma 4 for Local Use

Part 4: Global Multimodal and Open Source Models

Alibaba's Quinn 3.5 Omni and 3.6 Plus: Multimodal and Agentic Models

RC's Trinity Large Thinking: A New Open Source LLM

Part 5: Integration, Acquisitions, and Productivity Tools

Claude's Computer Use in Code, Codex Plugin, and ChatGPT in CarPlay

OpenAI Acquires TBPN: A Tech Business Production Network

Perplexity's Computer for Taxes and Slack's New AI Capabilities

Part 6: Industrial and Retail AI Applications

GM's Use of AI to Visualize Cars and Speed Up Design

AI in Shopping Carts: Instacart's Smart Carts for Real-Time Influence

AI News: Anthropic Leak Shows Us The Future of AI

Matt Wolfe

Part 1: Anthropic's Code Leak and Memory Architecture

00:04Anthropic's Claude Code Leak: Insights into Memory Architecture and Kairos

Anthropic's Claude Code Leak: Insights into Memory Architecture and Kairos

01:15Kairos: Anthropic's Proactive AI Agent for Autonomous Task Execution

Kairos: Anthropic's Proactive AI Agent for Autonomous Task Execution

05:06Anthropic's Response to the Leak and Potential Claude Code Spinoffs

Anthropic's Response to the Leak and Potential Claude Code Spinoffs

Part 2: Creative Assets and OpenAI's Strategic Moves

06:31ReCraft V4: AI Models for Professional-Grade Visual Assets

ReCraft V4: AI Models for Professional-Grade Visual Assets

08:37OpenAI's Massive Raise and Plans for a Unified AI Super App

OpenAI's Massive Raise and Plans for a Unified AI Super App

10:18OpenAI's Agentic Focus, Sora Shutdown, and Microsoft's New Transcription Model

OpenAI's Agentic Focus, Sora Shutdown, and Microsoft's New Transcription Model

Part 3: Microsoft and Google Model Updates

11:43Testing Microsoft's MAI Transcribe 1 and Availability in Microsoft Foundry

Testing Microsoft's MAI Transcribe 1 and Availability in Microsoft Foundry

13:42Google's VO 3.1 Lite and AI Inbox for Ultra Subscribers

Google's VO 3.1 Lite and AI Inbox for Ultra Subscribers

15:25New Open Source LLMs: Google's Gemma 4 for Local Use

New Open Source LLMs: Google's Gemma 4 for Local Use

Part 4: Global Multimodal and Open Source Models

17:15Alibaba's Quinn 3.5 Omni and 3.6 Plus: Multimodal and Agentic Models

Alibaba's Quinn 3.5 Omni and 3.6 Plus: Multimodal and Agentic Models

19:15RC's Trinity Large Thinking: A New Open Source LLM

RC's Trinity Large Thinking: A New Open Source LLM

Part 5: Integration, Acquisitions, and Productivity Tools

21:55Claude's Computer Use in Code, Codex Plugin, and ChatGPT in CarPlay

Claude's Computer Use in Code, Codex Plugin, and ChatGPT in CarPlay

23:33OpenAI Acquires TBPN: A Tech Business Production Network

OpenAI Acquires TBPN: A Tech Business Production Network

24:27Perplexity's Computer for Taxes and Slack's New AI Capabilities

Perplexity's Computer for Taxes and Slack's New AI Capabilities

Part 6: Industrial and Retail AI Applications

26:20GM's Use of AI to Visualize Cars and Speed Up Design

GM's Use of AI to Visualize Cars and Speed Up Design

27:31AI in Shopping Carts: Instacart's Smart Carts for Real-Time Influence

AI in Shopping Carts: Instacart's Smart Carts for Real-Time Influence