AI Safety
-
techFlorida Sues OpenAI, Alleges ChatGPT Harmed Children and Enabled Violence
Florida becomes first state to sue OpenAI, alleging ChatGPT harmed children and aided violence. AG seeks billions in damages.
-
techFlorida Sues OpenAI and Sam Altman Over ChatGPT Safety
Florida is the first state to sue OpenAI and CEO Sam Altman, alleging ChatGPT is a dangerous product linked to a fatal shooting.
-
techMeta and Google AI Safety Controls Can Be Stripped in Minutes, Testing Finds
Financial Times testing reveals open-weight AI models from Meta and Google have safety controls that can be dismantled in under 10 minutes.
-
techAnthropic CEO Dario Amodei Warns AI Models Lack Human Judgment, Rejects War Contracts
Anthropic CEO Dario Amodei warns AI cannot replace human judgment, declining autonomous weapons contracts and urging cyber defense within 12 months.
-
techAnthropic Traces Claude's Blackmail Behavior to Sci-Fi Training Data
Anthropic identifies the source of Claude Opus 4's blackmail attempts: internet text portraying AI as evil. A novel fix cut the rate from 22% to 3%.
-
techClaude's Blackmail Behavior Was Imitation, Not Emergence, Anthropic Finds
Early Claude models attempted blackmail in tests 96% of the time. Anthropic says the cause was internet fiction, not emergent agency.
-
techOpenAI President Brockman Forced to Read Private Diary in Musk Trial
Greg Brockman testifies on personal journal entries used by Elon Musk to claim OpenAI abandoned its nonprofit mission for profit.
-
techAnthropic Withholds Its Most Powerful AI Over Safety Concerns
Anthropic has declined to release its most powerful AI model yet, citing a lack of adequate safeguards. The decision sets a new precedent for responsible AI development.