Google accused of using novices to fact-check Gemini's AI answers

There's no arguing that AI still has quite a few unreliable moments, but one would hope that at least its evaluations would be accurate. However, last week Google allegedly instructed contract workers evaluating Gemini not to skip any prompts, regardless of their expertise, TechCrunch reports based on internal guidance it viewed. Google shared a preview of Gemini 2.0 earlier this month.  

Google reportedly instructed GlobalLogic, an outsourcing firm whose contractors evaluate AI-generated output, not to have reviewers skip prompts outside of their expertise. Previously, contractors could choose to skip any prompt that fell far out of their expertise — such as asking a doctor about laws. The guidelines had stated, "If you do not have critical expertise (e.g. coding, math) to rate this prompt, please skip this task."

Now, contractors have allegedly been instructed, "You should not skip prompts that require specialized domain knowledge" and that they should "rate the parts of the prompt you understand" while adding a note that it's not an area they have knowledge in. Apparently, the only times contracts can skip now are if a big chunk of the information is missing or if it has harmful content which requires specific consent forms for evaluation. 

One contractor aptly responded to the changes stating, "I thought the point of skipping was to increase accuracy by giving it to someone better?" 

Google has not responded to a request for comment. 

This article originally appeared on Engadget at https://www.engadget.com/ai/google-accused-of-using-novices-to-fact-check-geminis-ai-answers-143044552.html?src=rss https://www.engadget.com/ai/google-accused-of-using-novices-to-fact-check-geminis-ai-answers-143044552.html?src=rss
Creato 5mo | 19 dic 2024, 15:10:13


Accedi per aggiungere un commento

Altri post in questo gruppo

How to watch the Android Show ahead of Google I/O 2025

Google's annual I/O developer conference is coming on May 20, and for the first time, there's two major events you'll want to watch to stay on top of all the updates the company's making to its sof

9 mag 2025, 04:40:14 | Engadget
ChatGPT Deep Research can now connect to GitHub

ChatGPT is bringing its Deep Resear

8 mag 2025, 23:50:19 | Engadget
Microsoft says it hasn't raised Surface prices

Microsoft hasn't secretly raised Surface prices, as earlier reports claimed. Instead, it has removed the base models of the Surface Pro 13-inch and Surface Laptop 13.8-inch from Microsoft.com, acco

8 mag 2025, 23:50:18 | Engadget
GoldenEye 007 and Quake join the World Video Game Hall of Fame

The World Video Game Hall of Fame welcomed its 2025 inductees today. The Strong National Museum of Play

8 mag 2025, 23:50:18 | Engadget
Meta will test video ads on Threads

Instagram's Threads app began

8 mag 2025, 21:40:15 | Engadget
The best last-minute Mother's Day gift: Gadgets and subscriptions mom will love

It's getting down to the wire to snag a Mother's Day gift that will arrive on time. But luckily, as of this writing, more than a few of these gifts will arrive before Sunday for Amazon Prime member

8 mag 2025, 21:40:14 | Engadget
Palworld removes Pal gliding as it continues its legal battle with Nintendo

Nintendo's lawyers have killed another Palworld

8 mag 2025, 21:40:13 | Engadget