Anthropic seems to understand that making certain topics off-limits for Fable 5 is something of a double-edged sword. The company writes that “the same queries that are beneficial in the hands of cybersecurity professionals and biology researchers could be dangerous if available to malicious actors.” That puts Anthropic in the somewhat awkward position of having to judge who is and is not trustworthy enough to have access to a model that it says has potentially dangerous capabilities. The company says it will be periodically expanding its existing Project Glasswing program “in consultation wit
Anthropic says these topics are too dangerous to let its Fable 5 model talk about… The ChatGPT Work rollout comes alongside the debut of the new GPT-5.6 model , a new model that OpenAI promises provides “stronger performance per dollar” for “your hardest work.” That model can operate in three tiers, with the most expensive and high-performance costing $5 for a million input token… …
… The company, which is known for its ChatGPT and Codex services and harnesses, hopes to ultimately own the full stack behind its models and products, reducing dependence on outside companies like Nvidia and ostensibly providing better performance or efficiency thanks to vertical integration. …
… But testing from the UK’s AI Security Institute in recent months found that Mythos Preview performed similarly to OpenAI’s GPT-5.5 on a suite of Capture the Flag challenges, suggesting Mythos’ performance is not “a breakthrough specific to one model.” Among the usual raft of fair-to-middling benchm… …
… According to Doshi, the team made numerous improvements in pre-training with Gemini 3.5 Flash, but insights gleaned from how devs use Gemini models are really paying off. “With post-training, we’re really starting to unlock some of the value of the feedback we’re getting from users, for example, fr… …
… The system is also set up so that it’s model agnostic, allowing it to be switched over to better-performing models as AI systems evolve. …
… Researchers had also disabled some of the cyber classifiers that AI model providers built into the models to prevent misuse. …