"More than half of the accurate responses were 'ungrounded,' meaning they linked to websites that did not completely support the information they provided. This makes it challenging to check AI Overviews' accuracy."
"But with Gemini 3, Google's A.I.-generated answers were more likely to be ungrounded than when the system was based on Gemini 2."
"This study has serious holes."
A study by Oumi revealed that Google's AI Overviews have a 9% inaccuracy rate, despite 91% accuracy with Gemini 3. The study indicated that over 50% of responses were ungrounded, meaning they lacked proper sources to verify the information. This issue appears to be more prevalent with Gemini 3 compared to Gemini 2. Google responded to the study, claiming it has significant flaws. The findings raise concerns about the potential for misinformation affecting millions of users.
Read at Search Engine Roundtable
Unable to calculate read time
Collection
[
|
...
]