AIAnthropic accused of withholding its latest model as AISI’s ‘regulator’ label is challenged
The House of Commons committees account says an AI Security Institute response confirms Anthropic withheld its latest model from UK regulators. A commenter challenges the wording, saying AISI is not a regulator.
TLDR
On September 17, 2026, the House of Commons committees account said an AI Security Institute (AISI) response confirmed that Anthropic withheld its latest model from UK regulators. It said the response followed a September 9 letter from the Business, Innovation, Science and Trade Committee. A commenter quoting the announcement argued that AISI is not a regulator, welcoming its public responses but urging public institutions to describe its role accurately.
Combined views
4.5K
4 Sources, first seen 18h ago
Anthropic accused of withholding its latest model as AISI’s ‘regulator’ label is challenged
The House of Commons committees account says an AI Security Institute response confirms Anthropic withheld its latest model from UK regulators. A commenter challenges the wording, saying AISI is not a regulator.