Google will not sell Chrome, after all. All you need to know about latest antitrust ruling
In the early days of September, everyone in the tech world watched with bated breath as US District Judge Amit Mehta delivered his ruling in the case of the US Department of Justice (DOJ) vs. Google. And no, that’s not us being overly dramatic. If Mehta had sided fully with the DOJ, Google could’ve been forced to sell Chrome — a move that would’ve rocked the online advertising industry (and beyond). But despite the government’s bold demands, the judge went soft on Google, opting for lighter remedies instead.
Below we’ll break down the process, the ruling and what came out of it, let’s circle back to how it all started.
Google ruled a monopoly in search
In August 2024, a year before he spared Google from the harshest possible remedy, Mehta found that Google acted illegally to maintain a monopoly in the search engine market. Arguably, that alone wouldn’t have been a big deal. If anything it confirmed what we all already knew. Google Search dominates with around 90% market share, it’s the default on Chrome (the world’s most popular browser, also owned by Google), and it’s pre-set on Apple’s Safari thanks to a massive cash deal with Apple.
It was actually that money trail that Mehta zeroed in on. Google’s strategy to remain the default search engine across browsers and mobile devices (like Samsung phones) was backed by billions — and the numbers are staggering. In 2021 alone, Google admitted to spending about $26 billion on these agreements. That’s almost triple what it had been paying in 2014. And that spending wasn’t just about convenience, it was about cementing dominance.
The DOJ argued that Google built a search empire, then used it to squeeze advertisers, crush competition, and fund further improvements using anticompetitive profits. The proposed remedies ranged from mild to absolutely seismic. On one end, the DOJ called for an end to Google’s exclusive distribution deals — no more automatic default status on Samsung or Apple devices. But then they went nuclear: they wanted Google to divest from Chrome, forcing Google to sell off the very browser that helps funnel billions of users into its search engine every single day.
Taking Chrome out of Google’s hands wouldn’t just shift browser market share, it could upend the entire online ad ecosystem. But none of this has happened because Mehta chose a softer approach.
Breakthroughs in AI is a key argument for going easy on Google
When Mehta handed down his remedies, he pointed to the rise of generative AI as a sign that the market is evolving. OpenAI, Perplexity, Claude, and a handful of others have emerged as potential challengers in how people search for and interact with information. So, in his view, he could not just rule as if it was still 2020 and no AI revolution has happened.
While AI chatbots’ ability to replace search is still up for debate, Mehta cited that “tens of millions” of users are already turning to AI instead of traditional search engines, a shift that for him was apparently enough to go easy on Google. Mehta also hinted that if he decided in favour of more drastic measures, it could have hurt the web ecosystem as a whole.
However, it’s not that Google walked absolutely scot-free. Here’s what in the ruling:
Data sharing: Google has been ordered to share some of its search data with “qualified competitors,” specifically, parts of its search index and aggregated user interaction data. This is meant to help other providers build more competitive search experiences. Importantly, advertising data was declared off-limits, which protects one of Google’s biggest profit engines. Google, however, signalled that it would protest part of the ruling about the data sharing, saying that it has “concerns about how these requirements will impact our users and their privacy.”
No exclusivity: Google can no longer enter into exclusive contracts that lock in Google Search as the default on browsers and devices. However, deals with Apple and Samsung, which have Google Search pre-installed, will stand — albeit modified to ensure they are not technically exclusionary. While it’s not entirely clear how this will play out in practice, it’s our understanding that the status quo will likely be upheld. All other non-exclusive deals that involve payments are also allowed.
No breakup: The DOJ’s most dramatic request was firmly rejected. Judge Mehta called such a move “incredibly messy and highly risky,” hinting at what many were already thinking: Google might simply be too big, too embedded, to break up cleanly.
Our take: AI argument feels lukewarm, privacy risks are real
The judge leaned heavily on the idea that breakthroughs in generative AI are already reshaping the search landscape. And while there’s a chance that AI could eventually disrupt Google’s dominance, that’s far from guaranteed. If anything, there’s a very real possibility that Google, thanks to its existing monopoly in search, and its control over Chrome and Android, could simply carry its dominance into the AI space as well. Instead of disruption, we might just get consolidation.
Then there’s the issue of data sharing. On paper, it sounds like a reasonable step toward a more level playing field. The ruling says that Google is required to share index and interaction data subject to privacy safeguards. However, when we’re talking about big data on the scale Google operates, how those safeguards would actually work remains murky. This part of the remedy, while well-intentioned, could end up raising more questions than it answers — or worse, introducing new privacy risks without meaningfully boosting competition. Solving a competition problem at the expense of consumers’ privacy is unfair and might turn out deeply damaging.
Finally, Mehta’s biggest underlying concern seems to have been protecting the broader web ecosystem. He explicitly said that a forced breakup would be “incredibly messy and highly risky,” which, in our view, is the real reason the harsher remedies were taken off the table. But that raises a bigger question: is avoiding disruption enough to justify letting a dominant player stay dominant? We don’t know, but we don’t think it’s fair.
If “too big to fail” defined the post-2008 financial world, we might need a new term for Big Tech in the AI era: “too big to split.” And that’s exactly what this ruling feels like.
