Security & Safety
620,000 attacks expose a 35-point safety gap in reasoning models
TELUS Digital tested 34 AI models with more than 620,000 adversarial attacks. The benchmark shows why enterprise AI safety is now an operating discipline.
Models & Research
A 73% Agent Report Card Ends the Model-Only Benchmark Era
Open Agent Leaderboard evaluates full agent systems, not just standalone models, combining architecture, tools, cost, and failure behavior.
Models & Research
1,000 Desktop Tasks Turn Computer-Use Agents Into Verifiable Systems
OpenComputer shifts computer-use agent evaluation from LLM judges to reproducible desktop tasks and app-state verifiers.
Models & Research
The bill behind 73% success, agent evaluation moves beyond models
IBM Research and Hugging Face’s Open Agent Leaderboard evaluates AI agents as systems, including harnesses, costs, and failure modes.
AI Coding
Overeager Coding Agents Put Permission Boundaries on the Benchmark
OverEager-Bench measures whether coding agents cross the user’s authorized scope during benign tasks, using 500 scenarios and roughly 7,500 runs.
Security & Safety
Composer 2.5 shows Cursor training for reward hacking
Cursor Composer 2.5 shows the coding-agent race shifting from benchmark scores toward long-task failure points, targeted feedback, and reward-hacking detection.
Industry & Policy
Anthropic and Gates take Claude beyond the market
Anthropic and the Gates Foundation are pairing Claude credits, grants, connectors, datasets, and benchmarks for public-interest AI deployments.
Industry & Policy
Frontier AI predeployment review is becoming the new launch gate
CAISI is expanding predeployment evaluation work with Google DeepMind, Microsoft, and xAI, moving frontier AI launches beyond public benchmarks.