Show Notes
Has GPT-5.6 pushed today’s AI models as far as they can go? Frank and Justin explore OpenAI’s latest model, its striking real-world benchmark results, and whether its new computer-use tools can make everyday tasks like managing email and calendars dramatically easier. Plus: Anthropic claims it can inspect Claude’s hidden reasoning, raising new questions about misalignment, subconscious thought and whether AI consciousness should be taken seriously. And Tilly Norwood, the controversial AI-gen...
More from The AI Argument
AI is finding security holes that have been hiding in software for years. Justin thinks the rush to find and patch them will leave us safer. Frank worries about abandoned websites, rogue agents and whether every hole will ever get fixed. Plus: Anthropic’s public benefit fund and Elon Musk’s SpaceX share pledge raise the question of who should share out AI wealth. Unitree’s kung fu robot shows off its bodyguard moves, with cars getting a particularly rough time. 01:09 PHP vulnerabilities: ho...
Meta Muse Wonder and Google’s Gemini 4 Argon bring fresh competition to OpenAI and Anthropic. Justin’s impressed by the benchmarks, while Frank gives both models the same website brief to see how they perform in practice. Argon gets closer to what he wanted, while Muse keeps burning tokens after the job is done. Plus: a hidden recorder at OpenAI, supermarket fridges hit by hackers, and Britain’s AI security worries; AI job interviews where bots talk over the candidate; Justin’s cancelled Orb ...
OpenAI’s claim to 700 maths solutions comes with a big catch: someone still has to check them. Justin thinks mathematicians need to adapt and put AI to work. Frank worries that answers without understanding leave us with a dangerous gap. And while Justin is happy to patch mistakes in maths, he’s freaking Frank out with the idea of biology labs using AI solutions that still contain errors. Plus: A vibe-coded Adobe clone raises questions about who you trust to build and support your software, ...
OpenAI announced Dots at DevDay, an always-on personal AI assistant that works in the cloud. It can connect to your tools and data, then take work off your plate. With more easy-to-use AI agents now on the market, people can send them to email and phone companies on their behalf. Justin thinks this will force companies to automate support, while Frank worries about a costly future of bots talking to bots. Plus: Waju’s Fo is a personal AI assistant available in the EU, while Meta’s Muse raise...
Meta’s Muse is an easy-to-use AI agent that lets people who aren’t techies book trips, shop, and get things done. Justin also thinks it could help small local sellers compete with Amazon. Frank worries about what Meta could learn about its users—and whether shops would lose their direct relationship with customers. Plus: Claude 5.5 and GPT-6 get cheaper, while ChatGPT’s model picker frustrates Frank. Google tests a GPU in space, Anthropic sets up a wet lab, and Trump declares the age of super...