Psychological methods reveal major weaknesses in AI security testing
Researchers at the UK AI Security Institute used psychometric methods to show that popular safety benchmarks for language models don't measure one consistent trait. Blanket blocking of requests can artificially inflate a safety score even as the model gets less useful day to day. The study also offers a method for catching models that act more cautious during tests than they do in normal use.<br /> The article Psychological methods reveal major weaknesses in AI security testing appeared first on The Decoder. [...]
Rating
Innovation
Pricing
Technology
Usability
We have discovered similar tools to what you are looking for. Check out our suggestions for similar AI tools.
Red teaming LLMs exposes a harsh truth about the AI security arms race
Unrelenting, persistent attacks on frontier models make them fail, with the patterns of failure varying by model and developer. Red teaming shows that it’s not the sophisticated, complex attacks tha [...]
AWS Continuum integrates with OpenAI Codex and Anthropic Claude Code in major AI security push
Amazon Web Services is threading its AI-powered security infrastructure directly into the coding environments built by two of its fiercest rivals — and in doing so, it is making a bold bet that cont [...]
Intent-based chaos testing is designed for when AI behaves confidently — and wrongly
Here is a scenario that should concern every enterprise architect shipping autonomous AI systems right now: An observability agent is running in production. Its job is to detect infrastructure anomali [...]
Anthropic vs. OpenAI red teaming methods reveal different security priorities for enterprise AI
Model providers want to prove the security and robustness of their models, releasing system cards and conducting red-team exercises with each new release. But it can be difficult for enterprises to pa [...]
AI persuades best by overwhelming people with information instead of using psychological tricks
A new study challenges the idea that AI's persuasive power comes from personalization or psychological tricks. Instead, researchers found that simply overwhelming people with information—even w [...]
Anthropic and OpenAI just exposed SAST's structural blind spot with free tools
OpenAI launched Codex Security on March 6, entering the application security market that Anthropic had disrupted 14 days earlier with Claude Code Security. Both scanners use LLM reasoning instead of p [...]
MCP stacks have a 92% exploit probability: How 10 plugins became enterprise security's biggest blind spot
The same connectivity that made Anthropic's Model Context Protocol (MCP) the fastest-adopted AI integration standard in 2025 has created enterprise cybersecurity's most dangerous blind spot. [...]
GitHub leads the enterprise, Claude leads the pack—Cursor’s speed can’t close
In the race to deploy generative AI for coding, the fastest tools are not winning enterprise deals. A new VentureBeat analysis, combining a comprehensive survey of 86 engineering teams with our own ha [...]