- AI Platforms & Assistants
Google's latest AI model would rather bluff than admit confusion
When you purchase through links on our site, we may earn an affiliate commission. Here’s how it works.
(Image credit: Google)
Share
Share by:
- Copy link
- X
- Threads
- Gemini 3 Flash often invents answers instead of admitting when it doesn’t know something
- The problem arises with factual or high‑stakes questions
- But it still tests as the most accurate and capable AI model
Gemini 3 Flash is fast and clever. But if you ask it something it doesn’t actually know – something obscure or tricky or just outside its training – it will almost always try to bluff its way through, according to a recent evaluation from the independent testing group Artificial Analysis.
It seems Gemini 3 Flash hit 91% on the “hallucination rate” portion of the AA-Omniscience benchmark. That means when it didn’t have the answer, it still gave one anyway, almost all the time, one that was entirely fictional.
AI chatbots making things up has been an issue since they first debuted. Knowing when to stop and say I don't know is just as important as knowing how to answer in the first place. Currently, Google Gemini 3 Flash AI doesn’t do that very well. That's what the test is for: seeing whether a model can differentiate actual knowledge from a guess.
You may like-
Think you can trust ChatGPT and Gemini to give you the news? Here's why you might want to think again
-
I tested Gemini 3 Flash against Google Search
-
Google launches Gemini 3 Flash — and claims it's as fast as using traditional Search
Lest the number distract from reality, it should be noted that Gemini’s high hallucination rate doesn’t mean 91% of its total answers are false. Instead, it means that in situations where the correct answer would be “I don’t know,” it fabricated an answer 91% of the time. That’s a subtle but important distinction, but one that has real-world implications, especially as Gemini is integrated into more products like Google Search.
Ok, it's not only me. Gemini 3 Flash has a 91% hallucination rate on the Artificial Analysis Omniscience Hallucination Rate benchmark!?Can you actually use this for anything serious?I wonder if the reason Anthropic models are so good at coding is that they hallucinate much… https://t.co/b3CZbX9pHw pic.twitter.com/uZnF8KKZD4December 18, 2025
This result doesn't diminish the power and utility of Gemini 3. The model remains the highest-performing in general-purpose tests and ranks alongside, or even ahead of, the latest versions of ChatGPT and Claude. It just errs on the side of confidence when it should be modest.
The overconfidence in answering crops up with Gemini's rivals as well. What makes Gemini’s number stand out is how often it happens in these uncertainty scenarios, where there’s simply no correct answer in the training data or no definitive public source to point to.
Hallucination Honesty
Part of the issue is simply that generative AI models are largely word-prediction tools, and predicting a new word is not the same as evaluating truth. And that means the default behavior is to come up with a new word, even when saying "I don't know" would be more honest.
Get daily insight, inspiration and deals in your inboxContact me with news and offers from other Future brandsReceive email from us on behalf of our trusted partners or sponsorsBy submitting your information you agree to the Terms & Conditions and Privacy Policy and are aged 16 or over.OpenAI has started addressing this and getting its models to recognize what they don’t know and say so clearly. It’s a tough thing to train, because reward models don’t typically value a blank response over a confident (but wrong) one. Still, OpenAI has made it a goal for the development of future models.
And Gemini does usually cite sources when it can. But even then, it doesn’t always pause when it should. That wouldn’t matter much if Gemini were just a research model, but as Gemini becomes the voice behind many Google features, being confidently wrong could affect quite a lot.
There’s also a design choice here. Many users expect their AI assistant to respond quickly and smoothly. Saying “I’m not sure” or “Let me check on that” might feel clunky in a chatbot context. But it’s probably better than being misled. Generative AI still isn't always reliable, but double-checking any AI response is always a good idea.
Follow TechRadar on Google News and add us as a preferred source to get our expert news, reviews, and opinion in your feeds. Make sure to click the Follow button!
And of course you can also follow TechRadar on TikTok for news, reviews, unboxings in video form, and get regular updates from us on WhatsApp too.
TOPICS AI
Eric Hal SchwartzSocial Links NavigationContributorEric Hal Schwartz is a freelance writer for TechRadar with more than 15 years of experience covering the intersection of the world and technology. For the last five years, he served as head writer for Voicebot.ai and was on the leading edge of reporting on generative AI and large language models. He's since become an expert on the products of generative AI models, such as OpenAI’s ChatGPT, Anthropic’s Claude, Google Gemini, and every other synthetic media tool. His experience runs the gamut of media, including print, digital, broadcast, and live events. Now, he's continuing to tell the stories people want and need to hear about the rapidly evolving AI space and its impact on their lives. Eric is based in New York City.
Show More CommentsYou must confirm your public display name before commenting
Please logout and then login again, you will then be prompted to enter your display name.
Logout Read more
Think you can trust ChatGPT and Gemini to give you the news? Here's why you might want to think again
I tested Gemini 3 Flash against Google Search
Google launches Gemini 3 Flash — and claims it's as fast as using traditional Search
Thinking of taking a break from ChatGPT? Google Gemini is ready to impress
Should you use Google AI Mode or is boring old Search better?
I tested Gemini 3 and ChatGPT 5.1 head-to-head on what really matters
Latest in AI Platforms & Assistants
This gift-wrapping robot is quite funny, actually
ChatGPT’s new settings range from corporate calm to chaotic bestie
Claude's Chrome extension delivers convenience at the cost of privacy
A robot just learned 1,000 tasks in a single day — and it’s a big deal for everyday AI
Pinned chats in ChatGPT are here – and so is a mildly annoying restriction
Microsoft AI CEO admits Gemini 'can do things that Copilot can’t do'
Latest in News
Large External SSDs are now cheaper than internal ones as 4TB SATA SSD face extinction due to negligible price difference
Mullvad VPN boosts WireGuard speeds and stability with new Rust-based engine
HPE tells customers to patch OneView immediately as top-level security flaw spotted
Federal judge blocks Louisiana’s social media age verification law – here's why
Motherboards from Gigabyte, MSI, ASUS, ASRock at risk from new UEFI flaw attack - here's what we know
'We put the most pressure on ourselves' — Tomb Raider studio head on remaking one of the most iconic games of all time
LATEST ARTICLES- 1Gemini 3 Flash is smart — but when it doesn’t know, it makes stuff up anyway
- 2Watch out, Nvidia - Qualcomm acquires Alphawave Semi in latest addition to its AI data center push
- 3TechRadar Gaming's favorite gaming devices of 2025: personal picks from all the year's gear
- 4Large External SSDs are now cheaper than internal ones as 4TB SATA SSD face extinction due to negligible price difference
- 5Arm sheds billions in market capitalization after Qualcomm hints at RISC-V adoption with Ventara Micro acquisition