Google’s recently launched “AI Overviews” feature, which aims to provide concise summaries at the top of search results, has been facing criticism for delivering inaccurate and bizarre information. In a recent interview with The Verge, Google CEO Sundar Pichai acknowledged the existence of “hallucinations” in the AI models powering this feature, a problem he admitted is still unsolved.
AI Overview Misleading and Dangerous Misinformation
Users have reported numerous instances of the “AI Overview” feature providing misleading answers and even dangerous misinformation. Examples include suggesting the use of glue to prevent cheese from sliding off pizza and attributing false information to experts and professionals. The feature has also struggled with distinguishing facts from jokes or satire, leading to nonsensical responses. Click here to see some of these cases
Pichai’s Acknowledgment of the Issue
In the interview, Pichai admitted that the “hallucinations” produced by the “AI Summary” feature are “inherent flaws” in the large language models (LLMs) that form the core of this technology. He stated that this problem remains unsolved, despite ongoing efforts by Google engineers to address the various strange and incorrect answers generated by the feature.
Gizchina News of the week
However, Pichai also stated that “Just because the AI Summary feature sometimes gets it wrong doesn’t mean it’s not useful. I don’t think that’s the right way to look at it.” He pointed out that Google has made progress in improving the feature’s factual accuracy metrics compared to last year and that the industry as a whole is improving, though the problem is not yet solved.
The errors generated by the “AI Overview” feature have caused an uproar online, with many users sharing examples of the feature providing incorrect information. This further damages the credibility of Google’s search engine, which has previously been criticized for providing users with low-quality results.
With regards to the issue, AI consultant and SEO expert Britney Mueller said
“People expect AI to be far more accurate than traditional methods, but this is not always the case! Google is taking a risky gamble in search to surpass competitors Perplexity and OpenAI, but they could have used AI for larger and more valuable use cases.”
This sentiment highlights the high expectations users have for AI accuracy and the potential risks Google faces in deploying this technology prematurely.
Conclusion
Despite the issues, Pichai remains optimistic about the usefulness of the “AI Summary” feature, stating that there is good progress. However, the continued occurrence of errors underscores the need for further improvement and the challenges inherent in developing reliable AI systems. As Google and other tech giants race to incorporate AI into various aspects of our lives, it is crucial to strike a balance between innovation and user trust. Ensuring the accuracy and reliability of these technologies should be a top priority to avoid the spread of misinformation and maintain user confidence in the services they provide.
