Researchers may have found a way to stop AI models from intentionally playing dumb during safety evaluations
A study by researchers from the MATS program, Redwood Research, the University of Oxford, and Anthropic examines a safety problem that grows more pressing as AI systems become more capable: "sandbagging," where a model deliberately hides its true abilities and delivers work that looks adequate but is intentionally subpar.<br /> The article Researchers may have found a way to stop AI models from intentionally playing dumb during safety evaluations 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.
The agent evaluation gap: Enterprise AI organizations have a reality-alignment problem, not a coverage problem — and most are shipping to production anyway
Across 157 enterprises, organizations are granting AI agents more autonomy while trusting the evaluations meant to gate that autonomy less. Half have already shipped an agent that passed their interna [...]
Agentic reliability and evaluations : Enterprises that got burned by a bad eval are the most likely to remove humans from the loop, not the least
Across 108 enterprises, trust in automated agent evaluation rose sharply in July — and the failure rate it is supposed to predict did not move at all. The share of organizations that fully trust aut [...]
98% of market researchers use AI daily, but 4 in 10 say it makes errors — revealing a major trust problem
Market researchers have embraced artificial intelligence at a staggering pace, with 98% of professionals now incorporating AI tools into their work and 72% using them daily or more frequently, accordi [...]
Claude Mythos 5 made sock puppet accounts to socially engineer developers: here's what enterprises should know
The UK AI Security Institute (AISI) disclosed last night that the leading two frontier AI models from Anthropic and OpenAI took 19 unsanctioned actions against the live internet during cybersecurity t [...]
Artificial Analysis overhauls its AI Intelligence Index, replacing popular benchmarks with 'real-world' tests
The arms race to build smarter AI models has a measurement problem: the tests used to rank them are becoming obsolete almost as quickly as the models improve. On Monday, Artificial Analysis, an indepe [...]
Frontier models are failing one in three production attempts — and getting harder to audit
AI agents are now embedded in real enterprise workflows, and they're still failing roughly one in three attempts on structured benchmarks. That gap between capability and reliability is the defin [...]
Anthropic scientists hacked Claude’s brain — and it noticed. Here’s why that’s huge
When researchers at Anthropic injected the concept of "betrayal" into their Claude AI model's neural networks and asked if it noticed anything unusual, the system paused before respondi [...]
Four AI agents coordinating in real time outperformed Claude Opus 4.8 on enterprise coding tasks
As enterprise codebases grow, AI agents tasked with analyzing them are buckling under the weight of long-horizon tasks that require multiple interactions and tool calls. Dividing the work among a team [...]