Google adds AI to search box

- Google said on May 19 it is redesigning Search around Gemini-powered AI Mode, expanding the search box for longer, multimodal and follow-up queries. - Google said AI Mode now uses Gemini 3.5 Flash by default globally, while critics cite 60% zero-click searches and publisher traffic drops. - EU consumer groups filed complaints on May 21 against Google, Meta and TikTok over scam protections, according to Reuters.

Google said on May 19 that it was making the biggest change to its search box in more than 25 years, folding Gemini-powered AI deeper into Search and expanding the box to handle longer prompts. The update puts AI Mode closer to the front of Google’s core product, with support for text, images, files, video and Chrome tabs as inputs, according to the company. Google said Gemini 3.5 Flash is now the default model in AI Mode globally. The move comes as publishers warn that AI answers are reducing clicks to websites and regulators scrutinize how large platforms handle scams and manipulated content. ### How is Google changing the way people search? Google said at its I/O 2026 event that Search is being “completely reimagined with AI,” with a larger search box designed for more natural, paragraph-length prompts rather than short keyword strings. The company also said users can search with images, files, videos and Chrome tabs, broadening Search beyond typed queries. (blog.google) May 19 marked the point when Google also upgraded AI Mode to Gemini 3.5 Flash by default for users globally, according to the company’s Search update. Google has separately said AI Mode has been changing the kinds of questions people ask, with users entering longer and more complex queries in the United States. ### Why are publishers worried about this rollout? (blog.google) The New York Times reported that Google’s redesign could change how users reach information online because answers may be delivered more directly inside Search. That matters for publishers and other sites that depend on referral traffic from Google results. The Next Web said zero-click searches have reached 60% and that some publishers have seen traffic declines of as much as 89% as AI-generated answers absorb more user attention before any visit to an outside site. (blog.google) Those figures are part of a broader publisher complaint that Search is becoming more of a destination and less of a referral engine. (nytimes.com) Google has argued that its AI search features should still help people explore the web. In a May blog post, the company said it was improving how links appear in AI Mode and AI Overviews and developing new ways to help users find websites and sources they value. ### What does Google say AI Mode is supposed to do? Google said AI Mode is intended to handle more exploratory and multi-step questions than conventional search results. (ecommercenorthamerica.org) In its I/O announcement, the company paired the search-box redesign with broader agentic features in Search, part of a wider push to make Gemini act on behalf of users. (blog.google) Google also said in a May 19 post that AI Mode is changing user behavior in the United States by encouraging more nuanced questions and follow-ups. The company framed that as an expansion of Search rather than a replacement for the web, saying the product should help people discover a wider range of information. ### Where do trust and manipulation concerns come in? (blog.google) The BBC reported this week that Google and other AI companies have been responding to evidence that chatbot outputs can be manipulated into spreading misinformation. The report said Google updated its policies after researchers and journalists showed how AI systems could be poisoned or steered. (blog.google) Google has also said it is using AI in Chrome to block scams and improve online safety, though those product claims sit alongside broader questions about whether AI-generated answers can be trusted at scale. The concern for Search is that a more answer-driven interface gives users fewer reasons to inspect original sources for themselves. That is an inference based on Google’s product changes and the BBC’s reporting on manipulation risks. (vuink.com) ### Why are regulators paying attention now? Reuters reported on May 21 that EU consumer groups filed complaints against Google, Meta and TikTok, alleging they failed to protect users from financial scams on their platforms. The complaints expose the companies to possible regulatory action and add to pressure on large technology groups over consumer protection. (blog.google) May 21 is the latest concrete regulatory marker tied to this story because it shows scrutiny rising just as Google pushes AI deeper into the main search experience. Google’s next public reference point is its own Search product pages and post-I/O rollout updates, where the company is detailing AI Mode changes and link-display adjustments. (blog.google) (whbl.com)

Get your own daily briefing

Scout delivers personalized news, insights, and conversations tailored to your role and industry.

Download on the App Store

Shared from Scout - Be the smartest in the room.