The Future of Search and AI

Nick Fox - Google's SVP of Knowledge & Information
Watch the videoThe interview in a nutshell
In Google’s Nick Fox on the Future of Search and AI, Nick Fox – Google’s SVP of Knowledge & Information, overseeing Search, Maps, Commerce, and Ads – sits down with Semafor editor-in-chief Ben Smith at Google Marketing Live 2026. While much of the AI conversation assumes chatbots are quietly killing off search and the web, Fox makes the opposite case: search is entering its biggest redesign in 25 years, and AI is making the web more useful, not less. The conversation covers what’s changing for users, why Google didn’t rush its own AI response after ChatGPT, and what all of this means for marketers and shopping.
-
A Search Box Built for Real Questions: Users are no longer typing two-word keyword searches — they're showing up with multi-sentence problems, asking Search to understand what's actually in their head. Google's redesigned search box, unveiled at I/O the day before this interview, is built to expand and handle that complexity rather than force everything into a single line.
-
AI and the Web, Not AI vs. the Web: Fox pushes back on the narrative that AI will make the open web obsolete. His argument: AI is great for synthesis, but the web is where human experience lives — the reviews, the "here's what broke," the firsthand accounts people actually trust. Google's bet is that the best search experience combines both, though not everyone in the industry agrees the web comes out ahead in that arrangement.
-
Shopping Without the Friction: Conversational search (AI Mode) is reshaping how people discover what they want, while a new Universal Commerce Protocol (UCP) aims to remove the checkout friction of re-entering payment and shipping info every time. Together, Fox expects this to collapse the gap between "I'm curious about this" and "I just bought it."
Fox closes by framing this less as a single product launch and more as a long-term bet: Google’s reputation for accuracy – built over two decades – is the thing letting it move deliberately rather than chase every AI hype cycle. Whether that patience continues to pay off, or whether faster-moving competitors close the gap, is the open question hanging over the next phase of search.