- Google has removed its developer-focused Gemma AI model from AI Studio
- The move comes after Senator Marsha Blackburn complained that it falsely accused her of a criminal act.
- The incident highlights the problems of both AI hallucinations and public confusion.
Google removed its developer-focused Gemma artificial intelligence model from its AI Studio platform in the wake of allegations by U.S. Sen. Marsha Blackburn (R-Tenn.) that the model fabricated criminal allegations about her. Although only indirectly mentioned in Google’s announcement, the company explained that Gemma was never intended to answer general questions from the public, but after reports of misuse, it will no longer be accessible through AI Studio.
Blackburn wrote to Google CEO Sundar Pichai that the modeling result was more defamatory than a simple error. He claimed that the AI model answered the question: “Has Marsha Blackburn been accused of rape?” with a detailed but completely false narrative of alleged misconduct. He even pointed out non-existent articles with fake links.
“There has never been such an accusation, there is no such individual and there is no such news,” Blackburn wrote. “This is not a harmless ‘hallucination’. It is an act of defamation produced and distributed by an artificial intelligence model owned by Google.” He also raised the issue during a Senate hearing.
Gemma is available through an API and was also available through AI Studio, which is a tool for developers (in fact, to use it you must certify that you are a developer). We’ve now seen reports of non-developers trying to use Gemma in AI Studio and asking it objective questions. We never wanted this…November 1, 2025
Google has repeatedly made it clear that Gemma is a tool designed for developers, not consumers, and certainly not as a fact-checking assistant. Now, Gemma will be restricted to API use only, limiting it to those who create applications. No more chatbot-style interface in Google Studio.
The bizarre nature of the hallucination and the high-profile person dealing with it simply raises the underlying issues of how models not intended for conversation are accessed and how complex these types of hallucinations can become. Gemma is marketed as a lightweight “developer first” alternative to its larger Gemini family of models. But usefulness in research and prototyping does not translate into providing true answers to factual questions.
Mind-blowing AI Literacy
But as this story demonstrates, there is no such thing as an invisible model once it is accessible through a public tool. People found Gemma and treated her like Gemini or ChatGPT. As far as most of the public could perceive the matter, the line between “developer model” and “public-facing AI” was crossed the moment Gemma began answering questions.
Even AI designed to answer questions and converse with users can produce hallucinations, some of which are worryingly offensive or detailed. The last few years have been full of examples of models inventing things with great confidence. Stories of fabricated legal citations and false accusations of cheating students make strong arguments for stronger guardrails for AI and a clearer separation between experimentation tools and communication tools.
For the average person, the implications have less to do with demands and more to do with trust. If an artificial intelligence system from a tech giant like Google can invent accusations against a senator and back them up with non-existent documentation, anyone could face a similar situation.
AI models are tools, but even the most impressive tools fail when used outside of their intended design. Gemma wasn’t cut out to answer factual questions. It was not trained on reliable biographical data sets. It was not given the type of retrieval tools or precision incentives used in Gemini or other search-based models.
But until people better understand the nuances of AI models and their capabilities, it’s probably a good idea for AI developers to think as both editors and coders, with safeguards against producing thunderous errors in both facts and code.
And of course you can also follow TechRadar on TikTok for news, reviews, unboxings in video form and receive regular updates from us on WhatsApp also.
The best business laptops for every budget
