×
METR publishes cybersecurity assessment of leading AI models from Anthropic and OpenAI
Written by
Published on
Join our daily newsletter for breaking news, product launches and deals, research breakdowns, and other industry-leading AI coverage
Join Now

The Machine Ethics Testing and Research (METR) organization has completed preliminary evaluations of two advanced AI models: Anthropic’s Claude 3.5 Sonnet (October 2024 release) and OpenAI’s pre-deployment checkpoint of o1, finding no immediate evidence of dangerous capabilities in either system.

Key findings from autonomous risk evaluation: The evaluation consisted of 77 tasks designed to assess the models’ capabilities in areas like cyberattacks, AI R&D, and autonomous replication.

  • Claude 3.5 Sonnet performed at a level comparable to what human testers could achieve in about 1 hour
  • The baseline o1 agent initially showed lower performance but improved to match 2-hour human baseline performance after optimization
  • A specialized “elicited” agent scaffold was developed for o1, implementing an advisor-actor-rater loop that improved its performance

Testing limitations and caveats: Several factors prevented the team from establishing definitive capability boundaries.

  • The evaluation used default settings for o1’s reasoning parameters
  • Lack of access to o1’s internal thought processes limited capability assessment
  • Time constraints restricted the optimization of agent scaffolds
  • Incomplete information about the systems’ training methodologies affected testing approach

RE-Bench performance metrics: The evaluation included testing on RE-Bench, a suite of 7 complex AI R&D tasks.

  • Claude 3.5 Sonnet performed at the 37th percentile of human expert level
  • o1 achieved performance at the 30th percentile of human expert level
  • Different agent scaffolds showed varying levels of effectiveness for each model

Methodological challenges: The evaluation process revealed significant hurdles in capability-based safety assessment.

  • Small changes in agent scaffolding led to dramatic performance improvements
  • Performance scaled significantly with increased inference budgets
  • The need for extensive testing resources may make thorough evaluations prohibitively expensive

Future implications and research direction: While capability testing faces growing challenges, METR emphasizes its continued importance.

  • The organization plans to continue evaluating new AI systems, including the recently released DeepSeek-R1
  • Research into novel safety assessment methodologies remains a priority
  • The findings suggest a need for alternative safety arguments beyond pure capability testing

Complex evaluation landscape: The results highlight increasing difficulty in making definitive capability-based safety assessments for advanced AI systems, pointing to a need for more sophisticated evaluation methods and potentially alternative approaches to AI safety verification.

An update on our preliminary evaluations of Claude 3.5 Sonnet and o1

Recent News

AI agents transform workplace productivity

Companies are cutting task completion times in half by deploying specialized AI tools that work together to handle complex business processes.

Stanford research team finds why corporate AI projects succeed or fail

Clear executive ownership and smaller, focused teams prove more critical to AI success than technical capabilities alone.

AI-powered Perplexity unveils Deep Research for expert analysis

Perplexity's latest AI tool combines multiple data sources and cross-referencing to deliver research reports in minutes rather than hours.