Google Broke Antitrust Law — So Why Isn’t It Being Broken Up?
Google lost the antitrust case.
But it may have won the battle that mattered most.
A federal judge had already found that Google illegally monopolized key parts of the online advertising technology market.
The U.S. Department of Justice wanted a serious remedy:
force Google to sell part of its ad-tech empire.
On September 2, that request was rejected.
Google gets to keep AdX, the advertising exchange sitting at the center of the case. Instead of breaking the business apart, the court chose behavioral restrictions designed to make Google’s systems more open to competitors.
That creates an obvious question:
If a company illegally monopolized a market, why is it allowed to keep the business that helped create the monopoly?
The answer tells us a lot about how difficult it has become to regulate Big Tech.
First, What Did Google Actually Do Wrong?
This case was not mainly about Google Search.
It was about the invisible advertising machinery behind millions of websites.
When you open a news site and an ad appears beside an article, there is often an automated auction happening in a fraction of a second.
Publishers need software to manage their available ad space.
Advertisers need systems that place bids.
An ad exchange sits between them and helps match buyers with sellers.
Google operates at multiple points in that chain.
That vertical integration became the center of the government's case.
In April 2025, U.S. District Judge Leonie Brinkema found that Google had unlawfully monopolized two important markets:
publisher ad servers
and
ad exchanges for open-web display advertising.
The court also found that Google had improperly tied its publisher ad server to its AdX exchange.
That matters because publishers wanting access to important Google advertising demand had strong incentives to stay inside Google's ecosystem.
Think of Google as Owning Several Parts of the Auction House
The easiest way to understand the problem is with an analogy.
Imagine you want to sell something at auction.
One company:
provides the software you use to list the item,
owns the auction house,
brings many of the buyers,
helps decide which bids get considered,
and takes a fee when the transaction happens.
That company has a lot of control.
Now imagine competitors want access to that marketplace but cannot always compete on equal terms.
That is roughly why regulators became concerned about Google's ad-tech stack.
What Is AdX?
AdX stands for Google Ad Exchange.
It is the marketplace where advertising inventory can be sold through real-time auctions.
A publisher loads a webpage.
Different advertisers or advertising platforms compete for the available impression.
The winning ad is displayed almost instantly.
Reuters reports that publishers using AdX pay Google a 20% fee on those transactions.
It is an extremely technical business.
But it matters because online advertising is how a huge portion of the open web gets funded.
News sites.
Blogs.
Forums.
Specialist publications.
Free information services.
Many of them depend on this machinery.
The DOJ Wanted Google to Sell AdX
After winning the liability phase of the case, the Justice Department asked for something much more aggressive than a fine.
It wanted structural separation.
Specifically, regulators sought the divestiture of AdX and had also proposed separating Google's publisher ad-server business.
The logic was straightforward.
If Google's ownership of multiple layers helped create the anticompetitive structure, then separating those layers could remove the conflict permanently.
Behavioral promises can be violated.
A sold business cannot quietly be re-integrated later.
That is why antitrust regulators often prefer structural remedies in serious monopoly cases.
The Judge Said No
On September 2, Judge Brinkema declined to force Google to sell AdX.
Instead, she accepted most of the proposed behavioral remedies.
Those measures are intended to change how Google's ad-tech products interact with competitors rather than change who owns them.
One of the key ideas is greater access for competing systems to real-time bidding.
Other restrictions are aimed at preventing Google from unfairly favoring its own advertising services.
So Google keeps the business.
But it has to operate it differently.
Why Would a Judge Prefer That?
Because breaking up a complex technology system is much harder than saying:
“Sell this company.”
Google argued that AdX and its related ad infrastructure are deeply integrated into products used by publishers and advertisers around the world.
According to reporting on the ruling, Google said a forced separation could create:
technical disruption,
transition costs,
service instability,
and problems for customers who depend on the system.
The court evidently decided that less disruptive remedies could restore competition without forcing a major technical separation.
That is the theory.
Whether it works is another question.
This Is the Central Big Tech Antitrust Problem
Breaking up Standard Oil is conceptually easy to understand.
You divide physical businesses.
Digital platforms are different.
Products share:
data,
software,
APIs,
advertiser relationships,
identity systems,
cloud infrastructure,
machine-learning systems,
billing,
and enormous volumes of live transactions.
Separating one layer can be technically complicated.
That complexity can become a major advantage for dominant platforms.
The bigger and more integrated they become, the harder they are to dismantle.
Google Processes an Extraordinary Number of Ad Transactions
Google argued that its ad-tech system processes around 55 million requests per second.
That gives some sense of the scale involved.
A court-ordered breakup cannot simply turn off one server and turn on another.
Publishers around the world depend on the system.
Advertisers depend on it.
Agencies depend on it.
Entire businesses are built around the infrastructure.
So even after a court finds illegal conduct, the remedy itself creates risk.
But Critics See a Catch-22
This leads to a strange outcome.
A company becomes deeply embedded in an industry.
Its systems become difficult to replace.
Then regulators prove anticompetitive conduct.
But breaking it apart is considered too disruptive because so many businesses depend on it.
Critics argue that this effectively rewards scale.
The more essential a dominant company becomes, the harder structural enforcement gets.
That is one reason the ruling has frustrated antitrust advocates.
Behavioral Remedies Have a Mixed History
Behavioral remedies sound reasonable.
Tell Google:
do not preference yourself,
give competitors better access,
share certain information,
stop particular practices.
The problem is monitoring.
Technology changes quickly.
Auction systems are complicated.
Algorithms change.
New products appear.
Terms and interfaces can be modified.
Regulators then have to determine whether Google is technically complying while still achieving the same competitive advantage through some new mechanism.
That can become a years-long game of regulatory whack-a-mole.
Structural Remedies Are Simpler in One Important Way
Sell the exchange.
Different owner.
Different incentives.
That is easier to understand.
If Google no longer owned AdX, it would have less ability to design the exchange around the interests of its other advertising products.
That was the government's argument.
The DOJ essentially said:
Google's past behavior shows that rules alone may not be enough.
The judge disagreed that divestiture was necessary.
This Is Not Google's First Recent Antitrust Win on Remedies
And this is where the decision becomes much bigger.
Google has now avoided a forced asset sale in two major U.S. monopoly cases.
In its separate search monopoly case, the Justice Department sought the divestiture of Chrome.
That did not happen either.
Now regulators tried to force the sale of AdX.
That also failed.
Reuters described the new ruling as another symbolic defeat for U.S. efforts to use structural remedies against Big Tech.
Meta Also Avoided a Breakup
Google is not the only example.
U.S. regulators previously tried to unwind Meta's ownership of Instagram and WhatsApp.
That effort also failed to produce a structural breakup.
Together, these cases raise an uncomfortable question:
Can American antitrust law still realistically break up a modern technology platform?
Technically, yes.
Practically, it is proving extremely difficult.
Winning an Antitrust Case Is Only Half the Battle
This is the lesson most people miss.
There are really two cases.
Case 1: Did the company violate antitrust law?
The government can win this.
Case 2: What should the court do about it?
That can be much harder.
A monopoly finding does not automatically mean breakup.
The remedy must be justified, proportional and legally defensible.
And companies can argue that an aggressive remedy would harm customers or disrupt markets.
That is exactly where Google successfully fought back.
The Strange Part: AdX Is Not Even That Big Financially
You might assume AdX is one of Google's largest businesses.
It is not.
Court-related analysis cited by Reuters put Google's broader Ad Manager business, which includes AdX, at only around 4.1% of total revenue and 1.5% of operating profit in 2020. More recent figures have been redacted.
Financially, losing it would not destroy Alphabet.
Symbolically, though, it would have been enormous.
It would have shown that U.S. courts were willing to physically separate a major piece of a Big Tech platform.
That did not happen.
Why Publishers Should Care More Than Normal Google Users
Most consumers will notice absolutely nothing.
You will still use Google.
Gmail will still work.
YouTube will still work.
Search will still work.
But publishers have much more at stake.
The ad-tech market determines how much money publishers receive when advertisements appear on their websites.
If competition improves, publishers may potentially benefit from:
more bidding competition,
lower fees,
better transparency,
greater choice,
and less dependence on Google.
Whether the court's behavioral remedies actually create those improvements is now the important question.
This Matters to Independent Websites Too
The open web depends heavily on advertising.
Consider a small publication or blog.
It may spend months producing useful content.
The reader does not pay.
Instead, advertising funds the site.
Between the advertiser's budget and the publisher's revenue are multiple technology companies taking fees.
When one company controls several of those layers, the economics become important.
That is why this case is not just abstract competition law.
It affects who captures the value created by online content.
And Now AI Is Making the Situation Even More Complicated
Google's advertising monopoly fight is happening just as the internet's business model is changing again.
AI assistants are increasingly answering questions directly.
Google itself is integrating AI deeply into Search.
ChatGPT is building advertising products around conversational intent.
Publishers worry users may receive answers without clicking through to the websites that produced the underlying information.
So Google is simultaneously facing pressure over:
its traditional advertising infrastructure
and
its emerging AI search business.
Europe Is Already Looking at That Next Fight
The European Commission is currently questioning publishers about Google's AI Search policies, including its newly introduced option allowing publishers to exclude content from AI-generated summaries without losing traditional search visibility.
Publishers have argued that AI-generated answers could reduce traffic and advertising revenue.
That means regulators may soon be debating a newer version of the same old issue:
How much control should one company have over how information is discovered and monetized online?
This Is Where ChatGPT Becomes Relevant
The timing is fascinating.
Google is being ordered to make parts of its traditional advertising infrastructure more open.
At the same time, OpenAI is building a new advertising system around ChatGPT.
The two models are different.
Google historically monetized:
search queries.
OpenAI is trying to monetize:
conversational intent.
If users increasingly make purchasing decisions inside AI conversations, the future advertising market could look very different from the one Google currently dominates.
Google's Real Competition May Solve Part of the Antitrust Problem
There is an irony here.
Courts may hesitate to break up a company when technology itself is creating new competition.
AI search is already changing user behavior.
ChatGPT.
Perplexity.
Claude.
Gemini.
Other AI assistants.
If users increasingly bypass traditional search or traditional publisher advertising systems, Google's market power could weaken organically.
That possibility has already influenced thinking in other Google antitrust proceedings.
But waiting for technology to fix monopoly problems carries obvious risks.
New Technology Does Not Automatically Mean Competition
Google is also one of the world's biggest AI companies.
It owns Gemini.
It controls enormous cloud infrastructure.
It owns YouTube.
It controls Android.
It operates Search.
It has billions of users.
So AI disruption does not guarantee that Google loses power.
It could simply allow Google to become dominant in another layer of the internet.
That is exactly why regulators remain interested.
Google's Biggest Advantage Is Distribution
A startup can build a better product.
Google can place its product in front of billions of users.
That is incredibly powerful.
Chrome.
Android.
Search.
YouTube.
Gmail.
Maps.
All create distribution.
Modern antitrust cases increasingly revolve around that issue:
not simply whether a product is good,
but whether platform control makes it difficult for alternatives to compete.
So Did Google “Get Away With It”?
That framing is tempting.
But it is too simple.
Google did lose important parts of the case.
The court found illegal monopolization.
Its business practices will be restricted.
Competitors should receive better access.
Google will operate under court supervision.
Those are meaningful consequences.
But the company avoided the most dramatic consequence.
It keeps AdX.
That is why both sides can claim something resembling victory.
Google Still May Appeal
The story is also not necessarily finished.
Google has disputed the underlying monopoly findings and may continue challenging parts of the case.
The Justice Department has also said it is evaluating next steps after the remedies decision.
Large antitrust cases can continue for years.
So this should not be understood as the final end of Google's regulatory problems.
The Bigger Question Is Whether Courts Can Move Fast Enough
The ad-tech lawsuit was filed in 2023.
The underlying conduct stretches back much further.
The liability decision arrived in 2025.
The remedies decision came in 2026.
Technology markets can transform completely during that timeframe.
This is one of antitrust law's biggest weaknesses.
By the time a case ends, the market may already look different.
That's Especially Dangerous in AI
Imagine trying to litigate an AI monopoly case beginning today.
By the time:
discovery ends,
trial happens,
appeals finish,
and remedies are imposed,
the entire AI market could be different.
Models change every few months.
Companies appear and disappear.
Product categories merge.
That makes traditional antitrust enforcement extremely difficult.
Maybe the Real Answer Is Legislation
Some competition advocates argue that courts should not be expected to redesign digital markets one lawsuit at a time.
Congress could create clearer rules around:
interoperability,
self-preferencing,
data access,
platform conflicts,
and digital advertising markets.
Europe has already moved further in this direction through the Digital Markets Act.
The U.S. has largely relied more heavily on antitrust litigation.
The Google ruling demonstrates the limitations of that approach.
The Difference Between the U.S. and Europe Is Becoming Clearer
Europe increasingly tries to regulate dominant technology platforms before every practice becomes a decade-long antitrust case.
The U.S. more often sues after alleged harm has already occurred.
Both approaches have drawbacks.
Overregulation can slow innovation.
Litigation can arrive too late.
But Google's repeated success in avoiding structural separation may strengthen arguments for clearer ex-ante rules.
What Should We Watch Next?
There are four things that matter now.
1. The exact behavioral remedies
The detailed implementation will determine whether competitors genuinely gain useful access.
2. Publisher economics
Do publishers actually earn more or gain more bargaining power?
3. Competitor market share
Do alternative ad exchanges and ad servers grow?
4. Google's AI advertising strategy
The next monopoly battle may not involve today's ad-tech stack at all.
It may involve AI.
My Take: Google Won the Remedy More Than It Won the Case
That is probably the best way to understand what happened.
The government proved serious antitrust violations.
Google failed to erase that finding.
But when the moment came to decide what should actually happen to the company, Google avoided the remedy regulators wanted most.
No AdX sale.
No major structural breakup.
Instead, rules.
That is a significant victory.
Because ownership matters.
The Bottom Line
A federal court found that Google illegally monopolized important online advertising technology markets.
The Justice Department responded by asking the court to force Google to sell its AdX advertising exchange.
On September 2, Judge Leonie Brinkema refused.
Google will instead face behavioral restrictions intended to give competing ad-tech systems better access and prevent anticompetitive practices.
That does not mean Google escaped consequences.
But it does mean something significant.
For the second time, the U.S. government proved monopoly conduct involving Google and still failed to force the company to sell a major asset.
And that may be the bigger story.
The question facing regulators is no longer simply:
Can they prove Big Tech broke antitrust law?
They increasingly can.
The harder question is:
What can they realistically do about it once they win?
That answer remains much less clear.
Quick Facts
Related Reading
ChatGPT Ads Just Hit a $1 Billion Run Rate — Here’s What That Means for Users
The $3 Trillion Question: What’s Actually Being Built for the AI Boom
What Wall Street Is Actually Saying About AI Stocks Right Now
Meta Tried to Replace Workers With AI — Then the Plan Started Falling Apart
Comments
Post a Comment