5 different models dropped last week and the GPT-5.6 usage limits are brutal...
Summary
Theo and Ben review recent AI model releases, including Meta’s Muse Spark 1.1, OpenAI’s conversational GPT Live 1, and XAI’s highly praised Grok 4.5. They discuss the chaotic landscape of usage limits, the Apple-OpenAI lawsuit, and Elon Musk’s reactions. The episode also covers the controversial rebranding of the Codex app into the ChatGPT desktop experience and analyzes the engineering strengths of XAI versus OpenAI.
Topics discussed
Intro: AI model chaos and usage limits
OpenAI Live model: Voice capabilities and latency
GPT-5.5 and Codex Desktop app review
Grok Code Fast and xAI's competitive position
Industry landscape: OpenAI, Google, and Anthropic
OpenAI branding confusion: Codex vs. ChatGPT
OpenAI lawsuit against Apple and Sam Altman drama
Space data centers and underwater servers
Sponsor: WorkOS and MCP integration
GPT-5.6 models: Luna, Sol, and reasoning levels
Sub-agents, workflows, and token efficiency
Critique of Codex's system prompts and UI guidance
Using Cloud Code with proxy APIs and final thoughts
Listen ad-free on Castria