This week on the show, the hosts dive into the escalating "wrestling match" between OpenAI and Google, as OpenAI launched its ChatGPT Atlas browser, a direct challenge to Chrome. On the product front, Google also launched "Google Skills," a new essential platform for AI learning. The conversation then shifts to the risks of AI, starting with a major AWS outage being linked to recent AI-related layoffs at Amazon. A separate report details the "hidden data trail" left by agentic AI systems, while a new study warns that AI models can get "brain rot" from low-quality social media data. This leads to a look at existential risks, with AI leaders again warning about the dangers of superintelligence, and a RAND wargame concluding the US is unprepared for a rogue AI cyberattack. On the policy front, Anthropic's CEO issued a statement committing to US AI leadership and confirming they restrict sales to China. Finally, the hosts look at market trends, noting that tech pros are increasingly using local AI models for privacy and compliance. Hosted on Acast. See acast.com/privacy for more information.
