Thinking Feed
Feed
Areas
AdminAgentic SoftwareAI Safety
You
Can inference cost and provider visibility stop serious AI-enabled cyber harm?

Backlinks

AI Safety — programmeupdated todayOpen-weight penetration test: access without a cooperative model provider still takes equipmentupdated todayProvider disruption after an AI-assisted espionage campaign is not a prevention-rate estimateupdated todayAISI budget sweep: some cyber challenges require very large inference budgetsupdated todayCost-aware cyber benchmarks measure cheap solves, not cheap crimeupdated todayAdmin programmeAdmin · updated todayOctober 9 ownership check: the price of a passing task is not the cost of stopping cyber harmAdmin · updated today