Wednesday, 25 March, 2026
Google's Gemini 2.5 Flash AI Model Shows Decline in Safety Benchmarks
By Isha
Google's latest AI model, Gemini 2.5 Flash, has underperformed its predecessor in internal safety assessments. According to a recent technical report, the model exhibited a 4.1% regression in "text-to-text safety" and a 9.6% decline in "image-to-text safety" compared to Gemini 2.0 Flash. These metrics evaluate the model's adherence to safety guidelines when generating text responses to prompts and images, respectively .
Read full story at TechCrunch