Google has given users a clearer way to steer AI-generated news toward outlets they trust. Some 345,000 unique sites have already been added to the company's Preferred Sources setting, Google says. On 27 May 2026, Google expanded that setting so it now influences AI Overviews and AI Mode in Search, highlighting preferred outlets inside AI-generated summaries and responses. The rollout is live for English-language users worldwide, with a language expansion planned early next year.

It gives users a clear way to bias AI summaries toward trusted outlets, but it doesn't magically remove errors from AI-generated news.

That contrast is the practical read on Google's May 27 blog post, where the company announced Preferred Sources will now affect AI Overviews and AI Mode as well as regular Search, Discover, and Google News Top Stories. Google wrote that Preferred Sources "makes it easy to see more from your favorite websites," and that the label will help users "easily spot links in AI responses from the sources you've already selected." The company also said users have already chosen more than 345,000 unique sources, a figure it used to show uptake.

How to force your favourites into AI Overviews

The control is straightforward and available on mobile and desktop Search. To add a source you go to the preferences page, search by site name or address, and check the box for that site. When Google includes that site in an AI Overview it will be shown as a labelled link, the same way preferred outlets have already been identified in Top Stories and regular search results.

Practically, that means three things for anyone who cares where an AI summary finds its facts. First, mark the outlets you trust in Preferences. Second, expect those sites to appear more prominently when Google composes an AI Overview or when you use AI Mode. Third, follow the labelled links to the original reporting rather than relying solely on the AI-generated text.

Android Police reported that when someone picks a preferred source they click through to that site about twice as much on average. Google used that engagement metric to underline the feature's user benefit: preferred links are more visible and users tend to follow them at higher rates.

Mastheads

What else changed in AI Overviews

The Preferred Sources expansion was bundled with a set of refinements Google listed as five changes to AI Overviews.

Among them is a new section that surfaces direct quotes from public online discussions and social media, improved link previews that show creator or community names for social links, and prompts that suggest related topics after an AI Overview. Google also added a Highly Cited badge to more web article links, and said AI responses may indicate when an article explicitly references a Highly Cited source.

PCMag Australia showed an example where AI Overviews pulled quotes from a photography community and a subreddit in response to queries about photographing the northern lights, illustrating the social quote feature in action. Google said the social quote feature is meant to surface "helpful insights" from firsthand sources, and hover previews will give more context about where links originate.

Preferred Sources itself has a short documented history. Google first introduced it as an experiment in Google Labs and then launched it broadly in August 2025 for regular Search and Top Stories. The May 2026 change brings the capability into Google's AI features after initial tests with a subset of users in the United States and India. 9to5Google noted the company is working on using Preferred Sources as a ranking signal across AI features so a user's selections will show up more often, but that wider use is described as forthcoming rather than finished.

Google framed the move as a way to help people find "primary reporting that other articles are referencing" and to surface "helpful perspectives from online discussions, forums, and social media." For publishers, the practical upside is clearer placement in AI Overviews and higher click engagement from users who have marked them preferred.

Still, the change isn't without critics. Independent analyses and journalism groups have documented error rates and misrepresentations in AI news summaries, which is the motivation behind many of these tracing and labelling changes. SlashGear cited a joint study by the European Broadcasting Union and the BBC that found AI models "routinely misrepresent news content" with an error rate reported as high as 45 percent. Surveys from the Reuters Institute and Poynter show public scepticism about AI-produced news, and privacy and platform critics argue that a user-selected source list privileges established publishers and may not fully address AI hallucinations.

That is the trade-off built into the design. Preferred Sources gives users an explicit lever to nudge AI results toward outlets they trust, and Google is pairing that with more context in Overviews, such as Highly Cited badges and social quotes. But the company and outside commentators both signal that the change is an improvement in traceability rather than a cure for misstatements by AI systems.

For people who care which outlets feed into automated summaries, the step-by-step matter most. Add your preferred sites in Preferences, check that Google labels appear in the AI Overview, and click through to the original reporting. Expect the feature to behave the same across mobile and desktop Search, and to continue affecting Top Stories and other regular search experiences.

Related Articles

The Preferred Sources expansion began on 27 May 2026 for English-language users, with a broader language rollout planned early next year. Google is also testing using Preferred Sources as a ranking signal across AI features, and if it rolls that out more widely, being included could significantly raise a publisher's visibility in AI Overviews.

This article was created with AI assistance.