RegulationSep 25, 2026 · 01:392 sources
Anthropic Charges for Blocked AI Requests to Halt System Attacks
From marketbrief.now
Anthropic will now charge users for requests blocked by safety filters, including those related to distillation attacks, biology, and frontier LLM development, to prevent system probing. The policy applies to Claude Code, Claude.ai, and Cowork, with testing showing 99.7% of accounts did not trigger billable blocks and a false positive rate below 0.1%.
Read the sourceSign in to suggest
Sources
One ranked edition, every morning
The stories that matter today. Free, cancel anytime.
Telegram @aidr_today