74 stories from 15 sources
The week in AI and Technology
Click on a headline to see stories that shaped it.
Threads
6 threads
Anthropic first widened access to Claude Fable 5, then suspended tools over cyber risks as US officials moved to restrict its most capable models. By week's end, frontier systems delivered through the cloud faced the prospect of export-style controls.
Apple moved from unveiling a rebuilt Siri to presenting Apple Intelligence as a system-wide feature across its devices and apps. Follow-up reactions focused on Siri's more restrained behavior and early use on the Mac, reinforcing Apple's distribution-led strategy.
AI infrastructure investment expanded through Meta's Indian capacity deal and new efforts to raise or lower the cost of compute. At the same time, supply constraints, data-center heat, and resource use made local physical capacity a larger constraint on growth.
AI agents progressed from a proposed ChatGPT ecosystem into mortgage decisions, payment rails, consumer ordering, and industry-specific products. These moves show model providers and established platforms competing to control the workflows where agents act, not just the models behind them.
OpenAI began an IPO process as ChatGPT reached one billion monthly app users, while Google and OpenAI considered lower prices and Meta faced pressure to turn AI spending into revenue. The week paired extraordinary user growth with sharper demands for sustainable economics.
The week brought pressure on how AI systems use data, from Meta's expanded off-platform personalization to musician claims over training data and Canadian scrutiny of Grok's deepfake safeguards. State attorneys general also opened an inquiry into OpenAI's advertising and handling of sensitive information.
Day’s Recap