Crashes, No Sound, or Screen Glitches?
Random freezes, missing sound and display glitches usually trace back to one bad driver. Find and replace yours safely.Free scan · under a minutePC Slower Than It Used to Be?
A free scan shows the junk files, broken settings and background clutter dragging Windows down - then fixes them in one click.Free scan · Windows 10 & 11Partly—but the headline combines two different Google AI failures. In May 2024, Google Search vice president Elizabeth Reid acknowledged that some AI Overviews produced “odd, inaccurate or unhelpful” answers. Google and Alphabet CEO Sundar Pichai made his own “we got it wrong” admission in February 2024, but that statement concerned the Gemini app’s image generation and text responses, not Search.
What Google admitted about AI Overviews
AI Overviews are summaries generated inside Google Search. Google describes them as a customized language model working with Search’s ranking systems to identify relevant information and provide supporting links.
As an Amazon Associate I earn from qualifying purchases.
After bizarre examples spread online, Reid wrote on May 30, 2024: “But some odd, inaccurate or unhelpful AI Overviews certainly did show up.” That is the clearest admission tied to the Search feature—and it came from Google’s Search VP, not Pichai.
The genuine examples
- For the query “How many rocks should I eat?”, an Overview suggested that eating rocks could be beneficial. Google said the answer followed a satirical article republished on a software company’s website.
- An Overview advised using nontoxic glue to thicken pizza sauce. Google attributed that result to sarcastic or troll-like forum content.
These were failures of an answer system presenting questionable web material as useful guidance. They were not legitimate dietary or cooking recommendations.
#1 Best Overall
Not every viral screenshot was real
Google also said many widely shared screenshots were fabricated. It specifically denied that some dangerous-result images circulating online had appeared in actual AI Overviews. A viral screenshot therefore is not, by itself, proof that Google displayed the claimed answer.
Why the Search system produced bad answers
Google said the incidents exposed weaknesses in how the system handled queries and source material. Its explanations included:
- Misinterpreting a user’s wording or intent, especially with nonsensical queries.
- Failing to recognize satire, humor or deliberately absurd writing.
- Relying on user-generated content that offered misleading advice.
- Working with limited or poor-quality information on a topic.
Those are Google’s explanations, not an independent finding that the feature was generally reliable. A generated summary can sound confident while omitting the context that makes a source safe, accurate or obviously satirical.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
Independent evidence: the mushroom example
The Associated Press asked mycology and botany professor Mary Catherine Aime to review a mushroom-related AI Overview. AP reported that the answer omitted information she considered potentially dangerous. A later answer remained problematic in her assessment.
Rank #3
This example matters because the risk was not merely an absurd sentence. It showed how an answer can be hazardous through omission: leaving out identification limits, toxicity warnings or other context a reader needs before acting.
What Google changed after the backlash
Google said it made more than a dozen technical improvements and would continue monitoring feedback. The changes it described included:
- Improved detection of nonsensical queries.
- Better handling of satire and humor.
- Less reliance on user-generated content when it could produce misleading advice.
- Restrictions for queries where an Overview was not useful.
- Stronger protections for health-related searches.
Google also reported that it found a content-policy violation on fewer than one in every 7 million unique queries on which AI Overviews appeared. That is a Google-reported policy-violation measure, not an independently measured accuracy rate. It does not establish how often summaries were incomplete, misleading or simply wrong.
Recommended Free Tools
What Sundar Pichai actually admitted
Pichai’s widely quoted “we got it wrong” statement came on February 27, 2024, during the separate Gemini controversy. Gemini’s image-generation feature produced inaccurate or offensive depictions and, in other cases, became so cautious that it refused prompts it should have answered. Google said the feature used Imagen 2 and paused image generation of people while it addressed the problem.
In a statement reproduced by Semafor, Pichai said: “I know that some of its responses have offended our users and shown bias – to be clear, that’s completely unacceptable and we got it wrong.” He added: “We’ll be driving a clear set of actions, including structural changes, updated product guidelines, improved launch processes, robust evals and red-teaming, and technical recommendations.”
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Search AI Overviews and Gemini were different incidents
| Aspect | AI Overviews | Gemini controversy |
|---|---|---|
| Product | Google Search’s AI-generated summaries | Gemini app image generation and text responses |
| Date of public admission | May 30, 2024 | February 27, 2024 |
| Failure mode | Misleading, inaccurate or unhelpful summaries, including failures to recognize satire and risky omissions | Inaccurate or offensive images, biased responses and refusals caused by over-cautious tuning |
| Spokesperson | Elizabeth Reid, Google vice president of Search | Sundar Pichai, Google and Alphabet CEO |
| Announced response | More than a dozen technical changes, source restrictions and stronger safeguards | Structural changes, revised guidelines, improved launch processes, evaluations, red-teaming and technical recommendations |
What readers should conclude
The accurate version of the headline is narrower than “Google’s CEO admits its AI totally screwed up.” Google acknowledged real failures in Search AI Overviews, but Reid—not Pichai—made that admission. Pichai’s explicit “we got it wrong” language referred to Gemini’s separate image and text problems.
The episodes also show why an AI summary should not be treated as an authority, particularly for health, food safety or other high-stakes decisions. The available sources do not provide an independently measured present-day accuracy rate for AI Overviews, so the May 2024 policy-violation figure should not be presented as one.
Quick wins for a faster PC:
Clear out junk files and repair common Windows errorsFree Scan →Scan for outdated or missing drivers - takes under a minuteDriver Scan →Repair Windows errors before they cause bigger problemsFix Now →Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
