After the artificial intelligence-powered search tool returned what the firm refers to as “odd and erroneous” results to users’ internet searches, Google is making some modifications to its AI Overviews.
At Google’s yearly I/O developer conference last month, AI Overviews were unveiled. These days, a box of artificial intelligence (AI)-generated text with connections to external websites shows at the top of search results when users use Google Search to locate information on specific topics. The AI Overviews are accompanied by traditional search results, which is a significant change in Google’s information presentation.
In a blog post, Google Vice President Liz Reid stated that the AI Overviews feature is intended for situations in which a user wishes to “get both a quick overview of a topic and links to learn more.” The technology does this by employing the company’s large language model (LLM), Gemini.
“With an AI Overview what [Google] can really do is synthesize a lot of information and get you the answer that you’re looking for very quickly,” Joseph said.
Joseph explained that AI Overviews simplifies the process by giving visitors a succinct response after summarizing the relevant information, as opposed to bombarding them with pages of links to click through.
According to technology journalist and author of the book “How AI Ate the World: A Brief History of Artificial Intelligence – and Its Long Future,” Chris Stokel-Walker, users will experience less friction and won’t have to click through to multiple websites, which can be a pain if you just need information quickly.
But according to Stokel-Walker, the new function makes it more difficult for users of Google Search to confirm the veracity of the content they’re reading.
“We’ve kind of got used to over the last two decades of Google Search dominance to the results that we get to a search term being largely right,” he told ABC Audio. “Suddenly, if you get rid of that, as Google is proposing, and actually just shove an answer straight into the search results page that its created via generative AI, you have no real way of identifying and kind of analyzing that information to see whether it’s true or not.”
Concerns regarding the new feature also exist. One is that generative AI technology, both from Google and other companies, has come under fire for “hallucinating”—that is, producing false and misleading information.
In the few weeks that AI Overviews have been made public, for instance, Google Search has suggested to users that they should eat at least one small rock every day and that mixing glue into the tomato sauce is a good way to get cheese to stick to pizza—both of which are obviously very bad ideas. It also mentioned that in 2005, Andrew Jackson, the seventh president of the United States, who passed away in 1845, received his college degree.
According to Stokel-Walker, the final advantage of AI Overviews is determined by balancing the costs and conveniences. “You no longer have kind of click through five or six different pages and maybe several pages of search results to find the right answer, but also it does mean that either the answer might be wrong or it might not be the one that you actually want to get,” he stated.
“We’ve always been very clear about the limitations of LLMs, that there will be hallucinations sometimes,” stated Alex Joseph of Google. He added that for this reason, AI Overviews also provides citations to the websites it visits in order to produce its results.
“It’s part of the reason that we present all of the information for you holistically,” Joseph replied. “These are quick shortcuts to help you get some information to you quickly, but they’re followed along with areas where you can go, double check, verify.”
Joseph added, “We only show them on queries where we have a high confidence that it’s going to be helpful and actually enhance the experience.” Not all queries are best served by an AI Overview.
Following reports of strange reactions from certain social media users, Google declared that it had “more than a dozen technical improvements” to AI Overviews. These include restricting the amount of user-generated information and satirical or amusing webpages that are included in the data used to create AI Overviews, as stated in Liz Reid’s blog post. In relation to health material, Reid stated that Google has “launched additional triggering refinements to enhance our quality protections” and that the company “aim[s] not to show AI Overviews for hard news topics, where freshness and factuality are important.”
Furthermore, according to the blog post, “AI Overviews generally don’t ‘hallucinate’ or make things up in the ways that other LLM products might,” and the inaccurate responses are caused by “misinterpreting queries, misinterpreting a nuance of language on the web, or not having a lot of great information available.”
In addition to raising questions about accuracy, Stokel-Walker stated that Google giving AI Overviews precedence over conventional search results might have an impact on income and change the way people transact business online.
“Websites produce content; they try and make it attractive to Google. Google will show them in its search results. And as a result, people click through to their website, they then see ads off the back of that, and the publisher makes the money that allows them to put new content onto websites,” Stokel-Walker explained.
However, Stokel-Walker stated that websites may witness a decrease in traffic and consequently a decrease in ad revenue if AI-generated content were to take the top spot on the Google Search results page.
According to Stokel-Walker, the situation is ironic. This is due to the fact that Gemini, the LLM that Google use to produce its AI Overviews, depends on the pages that it currently appears before.
“These websites do still need to exist, and they do need to have a way of making income, because otherwise, there’s nothing for those AI-generated search results to be based on,” Stokel-Walker pointed out.
That being said, AI Overviews are only one of several features the company has planned for its line of technological goods, regardless of how the issues around them turn out in the end.
“I think that doing something like this as quickly as Google is doing it is concerning,” said technology journalist C. Scott Brown of the website Android Authority.
Plans for other tools akin to AI Overviews have been revealed by Google; these capabilities will attempt to respond to inquiries regarding particular webpages or YouTube films. According to Brown, the introduction of those elements will occur in the context of heightened competition.
“And the reason it’s doing that is because it feels like it has to. It has to keep up with companies – especially like OpenAI, for example – that are creating generative AI technologies that are threatening Google’s core business, which is delivering information to people through Google search, and thus delivering them advertisements that allow Google to make billions and billions of dollars,” said Brown.
“With Google seeing these things as a threat, it can’t just rest. It can’t figure out how to do this cleanly, and do it right,” Brown added. “It just has to do it.”