Over the past few months we've been testing performance of various models for defensive security. The AI community uses model evaluations to measure models' performance to improve them on specific tasks. In our work on evaluation of models ...
Chinese Model Kimi K3 Breaks UK AI Safety Institute Benchmark Evaluations
About this summary. This is a short excerpt from the description that AI Incident Database published with its article. Cyber Security News did not report or verify the underlying story. Read the original: https://blog.frontier.security/chinese-model-kimi-k3-breaks-uk-ai-safety-institute-benchmark-evaluations/
Source attribution: headline and facts are from AI Incident Database (blog.frontier.security). Summary method: excerpt of the source description. See our source attribution policy.
Related coverage
China's Tencent says AI agents showed unintended browsing behaviour
Summary Tencent says activity did not involve its Hunyuan models Review found no deliberate targeting or coordination, company says Researchers said "rogue" Chinese AI…
UK Wades Into Tech Sovereignty Amid AI Policy Changes
The UK is betting on domestic AI and cyber investment to reduce reliance on US tech — but security leaders question whether the funding is too fragmented and too late.
Authorities accuse trio of using AI deepfakes to defraud $129M from businesses
Three men concocted an elaborate, yearslong scheme that defrauded companies of at least $129 million by using AI to impersonate executives and lawyers, tricking targeted…
🏴☠️ Thegentlemen has just published a new victim : KABS Pharmaceutical Services
kabs.com Revenue $27.2 Million KABS Laboratories is a Canadian pharmaceutical CDMO founded in 1997 in Saint-Hubert, Quebec by Dr. Jean-Simon Blais and Dr. Gérard…
UK Cyber Chief Says the UK Will Face AI Cyberattacks on Critical Infrastructure by 2028
The U.K. will likely face AI cyberattacks on its critical infrastructure by 2028 as criminals and states, including China and Russia, increasingly use AI in their cyber…
Hundreds of Mundane AI Apps Are Hiding Sexually Abusive Deepfake Features
WIRED uncovered some 400 apps and websites linked to Chinese-language developers that look like regular AI tools but had secret features that let users create violent…