Get the latest tech news

One of Google’s recent Gemini AI models scores worse on safety


A recently released Google AI model scores worse on certain safety tests than its predecessor, according to the company's internal benchmarking.

These surprising benchmark results come as AI companies move to make their models more permissive — in other words, less likely to refuse to respond to controversial or sensitive subjects. The company claims that the regressions can be attributed partly to false positives, but it also admits that Gemini 2.5 Flash sometimes generates “violative content” when explicitly asked. Thomas Woodside, co-founder of the Secure AI Project, said the limited details Google gave in its technical report demonstrates the need for more transparency in model testing.

Get the Android app

Or read this on TechCrunch

Read more on:

Photo of Google

Google

Photo of safety

safety

Photo of scores

scores

Related news:

News photo

Google Plans to Roll Out Gemini A.I. Chatbot to Children Under 13

News photo

Justice Dept. Lawyers Say US Wants to Break Up Google’s Ad Technology

News photo

DOJ Asks Federal Judge to Break Up Google’s Ad-Tech Tools