METR Releases Independent Audit of Anthropic Pilot Sabotage and Agent Containment Risk Report
Frontier AI evaluation institute analyzes autonomous task execution benchmarks, sandboxed capability thresholds, and oversight…
2 min read
Business
Local reporting. The people, places and decisions behind the headlines.
Neighborhoods:
Topics:
All storiesSan FranciscoLocal BusinessCybersecurityRetail & ShoppingOpenAIBusinessModel Evaluation✕
Browse topics
- AdTech Security
- AI Acquisitions
- AI Agents
- AI Copyright Law
- AI Economics
- AI Governance
- AI Safety
- AI Training
- AI Workforce
- Airbnb
- Algorithm Design
- AlphaEvolve
- Anthropic
- Appellate Courts
- Arts & Culture
- Atomic Machines
- Autonomous Robots
- AVELAR
- Bay Area News
- Bay Area Tech
- Berkeley
- Blackwell Gpu
- BMW Group
- Broadway Plaza
- Business
- Calico Life Sciences
- Candiru
- Cardiology
- Cellular Aging
- Central Contra Costa
- Cerebras Systems
- Chamber of Commerce
- Character.AI
- Citizen Lab
- Civic Life
- Cloud Computing
- Cloud Infrastructure
- Cloud Security
- Co-Scientist
- Commercial Real Estate
- Community Events
- ConnectWise
- Contra Costa County
- Copyright Law
- Corporate Security
- Cybercrime
- Cybersecurity
- Dario Amodei
- Data Analytics
- Data Centers
- Dave Brehmer
- Deepmind Co Scientist
- DeepSeek
- Department of Homeland Security
- Department of Justice
- Design AI
- Developer Productivity
- Developer Tools
- DGX Spark
- Diablo Valley
- Digital Privacy
- Digital Publishing
- Digital Rights
- Dining & Food
- Dining & Retail
- Disney
- DMCA Section 1202
- Documentary Film
- Downtown Walnut Creek
- Drug Discovery
- Drupal
- East Bay Tech
- Edge AI
- Edge Computing
- Education
- Electronic Frontier Foundation
- Emeryville
- Employment law
- Enterprise AI
- Enterprise Cloud
- Enterprise Cybersecurity
- Enterprise Defense
- Enterprise IT
- Enterprise Software
- Executive Appointment
- F5
- Family
- FBI
- FDA Clearance
- Federal Budget
- Federal Deficit
- Federal Policy
- Figma
- Financial Services
- First Amendment
- Fiscal Policy
- Frontier Models
- Google DeepMind
- GPT-6 Astra
- GPT-6.1 Sol
- GPT-Live-1
- Guest Opinion
- Hardware
- Hex
- Industrial Next
- Infectious Disease
- Inference Efficiency
- Insider Threat
- IT Administration
- Life Sciences
- Local Business
- Macroeconomics
- Manufacturing Automation
- Masimo
- Medical AI
- Mercenary Spyware
- Meta
- METR
- Microsoft
- Microsoft 365
- Model Context Protocol
- Model Evaluation
- Multi-Agent Systems
- Municipal Infrastructure
- NGINX
- Ninth Circuit
- NVIDIA
- Open Models
- Open Source
- OpenAI
- OVHcloud
- Pleasant Hill
- Proofpoint
- Public Benefit
- Public Safety
- Quantum Computing
- Real-Time API
- Remote Code Execution
- Research Funding
- Retail & Shopping
- Robotics
- RTX AI
- Rtx Spark
- San Francisco
- San Ramon Valley
- Semiconductors
- Senior Living
- Silicon Valley
- Social Media
- Sports & Recreation
- Stanford Medicine
- Startups & Business
- Supply Chain Attack
- Task Substitution
- Tech Policy
- TechCrunch Disrupt
- Tenant Rights & Housing
- The Orchards
- Threat Intelligence
- Todos Santos Plaza / Downtown
- Transit & Streets
- US Treasury
- Varonis
- Voice AI
- Web Scraping
- Windows 11
- Zero-Day Vulnerability
Showing 1 story tagged Model Evaluation
Turn off filter ×Search and community filters apply to these 50 stories in this part of the archive. Continue to earlier reporting below.
Business