On Thursday, weeks after launching its strongest AI mannequin but, Gemini 2.5 Professional, Google revealed a technical report displaying the outcomes of its inner security evaluations. Nevertheless, the report is gentle on the small print, specialists say, making it tough to find out which dangers the mannequin would possibly pose.
Technical stories present helpful — and unflattering, at instances — data that firms don’t all the time broadly promote about their AI. By and enormous, the AI group sees these stories as good-faith efforts to assist unbiased analysis and security evaluations.
Google takes a distinct security reporting strategy than a few of its AI rivals, publishing technical stories solely as soon as it considers a mannequin to have graduated from the “experimental” stage. The corporate additionally doesn’t embrace findings from all of its “harmful functionality” evaluations in these write-ups; it reserves these for a separate audit.
A number of specialists TechCrunch spoke with had been nonetheless upset by the sparsity of the Gemini 2.5 Professional report, nonetheless, which they famous doesn’t point out Google’s Frontier Safety Framework (FSF). Google launched the FSF final yr in what it described as an effort to establish future AI capabilities that might trigger “extreme hurt.”
“This [report] may be very sparse, accommodates minimal data, and got here out weeks after the mannequin was already made accessible to the general public,” Peter Wildeford, co-founder of the Institute for AI Coverage and Technique, advised TechCrunch. “It’s unattainable to confirm if Google resides as much as its public commitments and thus unattainable to evaluate the security and safety of their fashions.”
Thomas Woodside, co-founder of the Safe AI Mission, mentioned that whereas he’s glad Google launched a report for Gemini 2.5 Professional, he’s not satisfied of the corporate’s dedication to delivering well timed supplemental security evaluations. Woodside identified that the final time Google revealed the outcomes of harmful functionality exams was in June 2024 — for a mannequin introduced in February that very same yr.
Not inspiring a lot confidence, Google hasn’t made accessible a report for Gemini 2.5 Flash, a smaller, extra environment friendly mannequin the corporate introduced final week. A spokesperson advised TechCrunch a report for Flash is “coming quickly.”
“I hope it is a promise from Google to start out publishing extra frequent updates,” Woodside advised TechCrunch. “These updates ought to embrace the outcomes of evaluations for fashions that haven’t been publicly deployed but, since these fashions may additionally pose severe dangers.”
Google might have been one of many first AI labs to suggest standardized stories for fashions, but it surely’s not the one one which’s been accused of underdelivering on transparency recently. Meta launched a similarly skimpy safety evaluation of its new Llama 4 open fashions, and OpenAI opted to not publish any report for its GPT-4.1 sequence.
Hanging over Google’s head are assurances the tech big made to regulators to keep up a excessive normal of AI security testing and reporting. Two years in the past, Google told the U.S. government it might publish security stories for all “important” public AI fashions “inside scope.” The corporate followed up that promise with similar commitments to other countries, pledging to “present public transparency” round AI merchandise.
Kevin Bankston, a senior adviser on AI governance on the Middle for Democracy and Expertise, referred to as the development of sporadic and imprecise stories a “race to the underside” on AI security.
“Mixed with stories that competing labs like OpenAI have shaved their security testing time earlier than launch from months to days, this meager documentation for Google’s high AI mannequin tells a troubling story of a race to the underside on AI security and transparency as firms rush their fashions to market,” he advised TechCrunch.
Google has mentioned in statements that, whereas not detailed in its technical stories, it conducts security testing and “adversarial purple teaming” for fashions forward of launch.