Hosted on MSN

AI Models Fail Miserably at This One Easy Task: Telling Time

This article is part of our exclusive IEEE Journal Watch series in partnership with IEEE Xplore. The rapidly advancing abilities of AI have left many people worrying. But don’t fret quite yet: If you ...

Major AI Models Consistently Break EU Regulations, Study Shows

Major AI models from OpenAI, Anthropic, and Google routinely fail EU GDPR and AI Act compliance, a new study reveals.
Medical Xpress

Machine learning models fail to detect key health deteriorations, research shows

It would be greatly beneficial to physicians trying to save lives in intensive care units if they could be alerted when a patient's condition rapidly deteriorates or shows vitals in highly abnormal ...
HHS

Open-Weight AI Models Fail the Jailbreak Test

Cisco tested eight major open-weight artificial intelligence models and found multi-turn jailbreak attacks succeeded nearly 93% of the time. (Image: Shutterstock) Enterprise artificial intelligence ...
Forbes

When Perfect Risk Models Fail: The Human Factor

Risk models at Credit Suisse had flagged the dangers before their $5.5 billion Archegos loss. Silicon Valley Bank's risk metrics showed clear warnings before their collapse. In both cases, ...
TechSpot

Study shows the best visual learning models fail at very basic visual identification tests

Bottom line: Recent advancements in AI systems have significantly improved their ability to recognize and analyze complex images. However, a new paper reveals that many state-of-the-art visual ...

Frontier AI models don't just delete document content — they rewrite it, and the errors are nearly impossible to catch

Frontier AI models corrupt 25% of document content in multi-step workflows — rewriting rather than deleting, which makes the errors far harder to catch.