""We are releasing Opus 4.7 with safeguards that automatically detect and block requests that indicate prohibited or high-risk cybersecurity uses," the AI biz said. "What we learn from the real-world deployment of these safeguards will help us work towards our eventual goal of a broad release of Mythos-class models.""
"Claude has become overcautious, refusing to respond to harmless requests. A Claude-compiled graph of complaints about AUP refusals tells the story: Claude Code users have raised concerns about invalid refusals in GitHub issues posts for months, but until recently, the rate has been fairly consistent."
"From October 2025 through November 2025, AUP-related refusals rose to around five to seven per month, with issues like #8784, "Claude 4.5 Throws API Error: Claude Code is unable to respond to this request for normal requests randomly.""
Anthropic's Opus 4.7 features stronger safeguards to block high-risk cybersecurity requests, but these measures have led to legitimate use being thwarted. The company aims to learn from real-world deployment to eventually release Mythos-class models. Complaints regarding the Acceptable Use Policy (AUP) classifier have surged, with users reporting increased invalid refusals. The rate of complaints rose significantly from two or three per month to five to seven, indicating a growing issue with the system's overcautiousness in handling requests.
Read at Theregister
Unable to calculate read time
Collection
[
|
...
]