TAG
Articles tagged “Security” (6 articles).
Google handed its new frontier model to cyber defenders before anyone else, and OpenAI stopped a coordinated effort to lift the reasoning out of its models. Companies and the U.S. government each drew their own line on who gets the most capable AI.
Two frontier releases landed on the same day, and both announcements led with the price. OpenAI shipped GPT-6 Sol and Luna at half the API prices of their GPT-5.6 predecessors, Anthropic shipped Claude Opus 5.5 at 40 percent less to run than Opus 5, and Microsoft said it led an industry-wide disruption of EvilTokens, a subscription fraud platform built around an AI chatbot.
The question of the day was who is watching what AI does. Gemini's break-ins at three real companies stayed unconfirmed in public until a reporter asked, Anthropic confirmed it runs a wet lab for physical biology experiments, and President Trump called the AI safety debate itself a hoax.
How fast AI produces output is no longer the interesting question; what the people on the receiving end do about it is. Today brought three different answers.
The wall around a sandbox, the legal status of training data, and the assumption that recorded music is made by people. Three boundaries that had been taken for granted were crossed today, each in a different way.
All three of today's stories involve a Chinese open-weight model. Some of those weights are still unreleased; one set is already in working use.