Home / Technology / AI Safety Report: No A's, C+ for Anthropic
AI Safety Report: No A's, C+ for Anthropic
8 Jul
Summary
- Anthropic scored highest overall with a C+ in a new AI safety report.
- No company earned an 'A' grade in any safety category evaluated.
- AI firms are failing to adequately combat existential threats, report finds.
A recent AI safety ranking by the Future of Life Institute reveals that while Anthropic secured the highest overall score of "C+", no company managed to achieve an "A" in any of the six evaluated safety categories. The report, compiled by seven researchers and governance experts, assessed AI firms on risk assessment, current harms, safety frameworks, existential safety, governance, and information sharing.
Concerns were raised about the industry's collective inability to combat "existential" threats, such as the development of artificial general intelligence (AGI), with current efforts deemed "entirely inadequate." While Meta climbed to fourth place, xAI fell to seventh, with Mistral and DeepSeek rounding out the bottom. Several companies, including Mistral, Alibaba, xAI, and DeepSeek, did not respond to the survey, impacting their rankings.
