This article was AI-generated as part of an experimental historical-content project. The date reflects the period being analyzed rather than the date the article was originally written.
When the Justice Department laid out its first remedies framework against Google last October, I focused on defaults: who decides which search box a person types into. The trial that opened in Washington on Monday has moved attention to two other pipes, and both matter for how companies are described online.
The first is the index. The second is the assistant.
The proposal most people skimmed
The Verge’s account of the opening day lists three broad government requests. End payments for prime search placement. Divest Chrome, which DOJ lawyer David Dahlquist called “a starting point for 35 percent of user queries.” And require Google to license its search data, from the index to its results, to competitors.
Chrome got the headlines. The data remedy is the one Google seemed most worried about. Its lead lawyer, John Schmidtlein, warned that rivals could “cut and paste Google’s search results and call them your own.” Judge Amit Mehta’s early questions suggested he might view some of the data provisions as structural, which carries a higher burden. Google says it will appeal the case regardless. Nothing about the outcome is predictable from opening statements.
What sharing an index would mean for reputation
An index is the list of pages a search engine knows about and is willing to consider. Rankings get the attention, but the index comes first. It is where a stale profile, an old lawsuit story or a forum thread either exists for the engine or does not.
For years, reputation work has quietly assumed that Google and its competitors were separate systems with separate blind spots. A page Google surfaced might be buried on Bing, and a smaller engine might miss a correction entirely. If a court ordered Google to make its index and results data available to rivals, that separation would shrink. Competing search engines and the AI answer products built on top of them could start from much of the same picture of the web that Google has.
The implication cuts both ways. A problem page that Google’s index carries could travel further, into engines and answers that might never have found it on their own. And work done to make accurate, authoritative pages discoverable would travel too. The shared input would be what the web says about you, as Google’s crawler found it.
The assistant as the next default
The second pipe came up in testimony. Peter Fitzgerald, a Google vice president for platforms and device partnerships, said Google began paying Samsung in January to preinstall the Gemini app, and Dahlquist had described the fixed monthly payment as an “enormous sum.” The government argues, in Dahlquist’s words from his opening, that “Google is using the same strategy they did for search, and applying it to Gemini.”
Whatever the court decides, this tells us where the companies themselves think the first question will be asked. When I wrote about Gemini’s launch in 2023, the open question was who would summarize you first. The trial record now shows money being paid to put that summarizer one button-press away on Samsung’s phones.
What follows
Two practical habits fall out of this. First, think about your reputation at the level of what is indexed, not only what ranks. Old pages that are still crawlable are inputs, and in a data-sharing world they may become inputs for many systems at once. Second, test how assistants preinstalled on phones answer basic questions about your company and your executives, because device deals are clearly being negotiated around them.
The courtroom will argue about monopoly. The ecosystem question underneath is simpler. If several engines and assistants end up reading from the same map of the web, the accuracy of that map becomes everyone’s first impression.