SECTION: Judge Amit Mehta could have dismantled Google. Instead, he left Chrome intact, imposing only a ban on exclusive distribution agreements and curbing limited search‑data sharing, as TechPolicy.press's analysis of the ruling details. That pattern—minimal structural fixes across the current wave of Big‑Tech cases—defines today's antitrust landscape.
Judge Amit Mehta had the option to break Google apart. He declined it, ordering Chrome to stay put and settling instead for a ban on exclusive distribution deals and limited search-data sharing, according to TechPolicy.press's analysis of the ruling. That choice, replicated across the current docket of Big Tech cases, is the story: American courts have a structural remedy available and are declining to use it.
The pattern holds across the board. Meta's monopolization case was dismissed outright once Judge Boasberg counted TikTok and YouTube as competitors, per the same TechPolicy.press analysis. Google's ad-tech liability finding from Judge Brinkema is still awaiting a remedy. Reuters reported that the Search ruling shows how tech can outpace antitrust enforcement, with the judge citing OpenAI's rise as evidence that markets, not courts, may discipline incumbents fastest.
What this means for boards:
No source in this reporting quantifies what any of this costs in margin. That number does not yet exist. The question to bring to the next board meeting is not what the fines will be, but what a permanent compliance and monitoring function costs annually, indefinitely.
The Google Search remedy is worth reading as a spec, not a headline. Judge Mehta's September 2025 order rejected the Chrome divestiture the DOJ wanted and imposed a narrower set of behavioral obligations instead: a ban on exclusive distribution contracts and a requirement to share limited search data with competitors, according to TechPolicy.press's January 2026 analysis. For anyone architecting around a platform dependency, that distinction changes the planning horizon. A structural remedy is an event; a behavioral remedy is a running condition, checked against contract terms and data-sharing APIs that someone has to build, version, and audit indefinitely.
That has a direct analogue in how DOJ cleared Google's Wiz acquisition. Rather than litigate the merger, Google made upfront commitments on cloud interoperability and third-party security tool access, and the deal closed on those terms, per Dechert's April 2026 note on vertical merger enforcement. The mechanism worth noticing: these aren't consent decrees negotiated after a finding of harm. They're commitments made before close, functioning like an interface contract the acquirer promises not to break.
FTC Commissioner Mark Meador's framing is the one worth pinning to an architecture review. He's described foreclosure in software ecosystems as often arising from "strategic design and contractual leverage," raising switching costs by design, rather than from any physical bottleneck, per Dechert's note. Teams shipping platform features that raise switching costs, non-portable data formats, API rate limits scoped to discourage multi-homing, bundled default integrations, should treat those choices as documented tradeoffs, not incidental implementation details. If a regulator's theory of harm is that leaving was made expensive on purpose, the changelog and the PRD become discovery material.
Courts have shown they won't order breakups: Meta's case was dismissed outright by Judge Boasberg in November 2025 once TikTok and YouTube were counted in the relevant market, per TechPolicy.press. That is the asymmetry that matters for planning. Courts will still order behavioral remedies, contract bans, data-sharing mandates, that have to be built and maintained rather than settled once. Google lost on adtech too. Judge Brinkema found monopolization of publisher ad servers and exchanges in April 2025; remedies were argued through that November and the decision is still pending as of early 2026, per the same analysis. Whatever remedy lands there will be another spec engineering has to implement, not a business unit engineering can forget existed.
Judge Amit Mehta had a divestiture order sitting in front of him and declined to sign it. The Department of Justice wanted Google to sell Chrome. What it got instead, in September 2025, was a set of behavioral constraints: no more exclusive distribution contracts, limited search data sharing with rivals, and Google's corporate structure intact (TechPolicy.press's January 2026 analysis). That outcome is now the pattern, not the exception, and it is worth being precise about what the pattern actually is before asking what it costs.
Google lost on search and got behavioral remedies. Google lost on adtech too: Judge Brinkema found monopolization of publisher ad servers and exchanges in April 2025, remedies were argued through November, and a decision was pending as of early 2026. The asymmetry that matters for planning is that courts have shown they won't order breakups. Meta's case was dismissed outright by Judge Boasberg in November 2025 once he concluded that, with TikTok and YouTube counted as substitutes, Meta lacks monopoly power in the market the FTC had defined (TechPolicy.press's January 2026 analysis). Three outcomes, one court system, roughly the same set of facts about platform dominance. The variance itself is a finding. Courts are not converging on a remedy theory; they are litigating each case as if the last one set no template.
The counterintuitive claim, and the one an allocator should sit with, is that behavioral remedies are not obviously a heavier burden than the structural remedies they replaced. A divestiture is a one-time balance sheet event with a clean before-and-after. A behavioral remedy is open-ended: contract terms banned indefinitely, data-sharing obligations that require ongoing engineering and legal review, and a court or monitor checking compliance for years, with no expiration date attached. DOJ sought the cleaner outcome and didn't get it. Google got the outcome with no fixed end date. Whether that trade favors Google's margins depends entirely on how expensive the monitoring turns out to be, and nothing in the public record from these cases prices that yet.
Judge Mehta's reasoning supplies a second finding, and it is arguably the more consequential one for how enforcement gets litigated going forward. Reuters reported that the ruling explicitly credited the emergence of generative AI competitors, OpenAI among them, as a check on Google's dominance that reduced the need for structural intervention (Reuters's September 2025 report). A market moving fast enough to plausibly discipline an incumbent before a court can is now, in one ruling, an argument against the remedy that would have addressed it. That doesn't require any assumption about how the technology performs. It only requires that the pace of AI product launches keeps outrunning the pace of litigation, which is a low bar given that antitrust cases run on a timescale measured in years and model releases run on one measured in months.
The regulatory picture only holds together, though, if it is read as antitrust-specific. It is not a broader story about a hardening federal posture toward tech. The SEC, over the same window, moved the opposite direction. Its April 2026 release on fiscal 2025 results counted 456 total enforcement actions and $17.9 billion in ordered monetary relief. That came against a backdrop of 1,095 investigated matters closed without any action, and the agency described its own recent history as a "course correction" away from prior years that had "misapplied" resources to "pursue media headlines and run up numbers" (SEC's April 2026 enforcement results release). Chairman Atkins framed the agency's crypto posture, branded "Project Crypto," around helping make the U.S. "the crypto capital of the world" (SEC's Cyber, Crypto Assets and Emerging Technology unit page). Commissioner Crenshaw dissented publicly, warning the agency had "dismissed SEC enforcement actions left and right, undermining the credibility of our lawyers and the agency overall." Treating antitrust's behavioral turn and the SEC's deregulatory turn as one undifferentiated "continuous regulation" trend would flatten two agencies moving in opposite directions into a single line that doesn't exist.
What does generalize is the theory of harm, not the intensity of enforcement. Ronald Wick at Cohen & Gresser argued that the current wave of Section 2 cases targets monopoly maintenance, conduct by firms already on top, rather than monopoly creation, and that the same framework is migrating toward healthcare and media (Cohen & Gresser's July 2026 analysis). FTC Commissioner Mark Meador has pushed the theory further inside tech itself, arguing that foreclosure now often runs through "strategic design and contractual leverage," switching costs engineered into a product, rather than any physical bottleneck. That framing is one DOJ used when it cleared Google's Wiz acquisition only after Google committed to preserving third-party cloud interoperability (Dechert's April 2026 note on vertical merger enforcement). No source in this record puts a dollar figure on what any of that compliance architecture costs. The margin question is real. It is also, as of this reporting, unanswered by anyone who has actually measured it.