Google Search improves snippets to prevent misinformation

The team behind Google Search is tweaking the featured snippets – the text boxes that sometimes spread false information while trying to provide help. The company has announced an update that should make the answers more accurate and avoid the problem of false premises, or questions where no definitive sounding answer would make sense. It’s linked to an extension in Google’s “about this result” option and low-quality warnings data invalidas well as a new partnership for information literacy lesson plans for middle and high school students.

Snippets appear under many searches, but because they seem to answer questions directly by quoting pages, they can: backfire in ways that standard question answers do not. In a presentation to reporters, Google gave some examples of these problems and how it is trying to solve them. For example, when you searched for how long it takes light to get from the sun to Earth, Google at one point offered a snippet that emphasized its distance from Pluto instead.

According to Search VP Pandu Nayak, the solution lies in finding consensus: facts that match in multiple top search results. Speaking to reporters, Nayak clarified that this consensus check comes from pages that Google has already rated as high quality, something Google hopes will be a snippet equivalent of Google bombing. “It doesn’t establish that something is reliable, it just looks at the top results,” says Nayak. But by looking at different pages Google already trusts and then looking for similarities, it hopes it can avoid highlighting the wrong details.

google

A separate issue is the “false premise” problem, a phenomenon where Google is trying to be a little bit at handy with fragments. If you’ve been asking a leading question for years about something that never happened, Google has often provided snippets that seem to confirm its fact, taking text snippets out of context from a semi-related page. For example, the search team’s example is “When did Snoopy kill Abraham Lincoln,” which at one point marked the date of Lincoln’s death in a fragment. Google calls these cases “not very common,” but it says it’s trained its systems to get better at detecting them and doesn’t offer a featured snippet at all, and it promises to cut the incidence of these inappropriate appearances by 40 percent. has reduced.

This doesn’t necessarily solve every snippet problem. Nayak acknowledged that neither system would help with an issue identified last year where Google de the exact opposite of good advice on dealing with attacks, listing a series of “don’ts” prohibitions as guidelines for what to do. “Things like that are really about making sure our underlying algorithms take enough out of context,” says Nayak, who says Google continues to make improvements that can avoid similar issues.

But the goal is to make snippets less likely to get confused and increase trust in search results, something emphasized by Google’s other changes. For about a year now, Google has been placing warnings above unreliable search results that can appear in current news situations. It now extends that to more general situations where it finds that there aren’t high-quality results for a query, and adds a piece of advice before people can scroll down the page to see the results. It doesn’t stop anyone from seeing content, but it ideally helps manage expectations about the reliability of the information.

Google is also expanding “About this page,” which shows you details about the website from which a particular result came. The option was previously available on Search, but is now launching in Google’s iOS app in English – you can swipe up while browsing the app to learn more details about it, which theoretically helps you to measure its reliability. The system will launch on Android later this year and in other languages ​​in the coming months.