A year ago, Google faced the prospect of being dismantled. Today, artificial intelligence (AI) and a new court judgment has helped it avoid this fate. Part of the reason is that AI poses a grave threat to Google’s advertising revenues.

“Google will not be required to divest Chrome; nor will the court include a contingent divestiture of the Android operating system in the final judgment,” according to the decision.

Google must share certain data with “qualified competitors” as deemed by the court. This will include parts of its search index, Google’s inventory of web content. Judge Mehta will allow Google to continue paying companies like Apple and Samsung to distribute its search engine on devices and browsers. But he will bar Google from maintaining exclusive contracts.

The history of this decision goes back to a 2024 ruling by federal judge Amit Mehta. It found that Google maintained a monopoly in the search engine market, notably by paying billions to companies including Apple and Samsung to set Google as the default search engine on their devices.

Almost a year later, the same US judge issued his final ruling, and the tone could not be more different. Google will not be broken up. There will be no choice screen on new phones.

The nature of the search engine market, where more users generate more data, and more data improves search quality, made it impossible for competitors to challenge Google, the court found in 2024.

The 2024 ruling itself was controversial. While high quality data enables a dominant firm to extract more profit from consumers, it also allows it to provide a better service. Decades of research in economics has shown that determining which effect is more important is not straightforward.

At the time, the US Department of Justice deemed the issue so serious that it considered breaking up Google as the only viable solution. For instance, it suggested forcing the company to sell its web browser, Google Chrome.

The government also proposed forcing device manufacturers to offer users a choice of search engines during set up, and compelling Google to share most of its data on user behavior and ad bidding, where advertisers compete in auctions to get their ads shown to users for a specific search query or audience. These so-called “remedies,” measures Google would be required to implement to end its monopoly, aimed to restore competition.

Limited sharing

So, what has changed in a year to so radically change the perception of Google’s market dominance? The main answer is AI—and specifically, large language models (LLMs) like ChatGPT, Claude, and Google’s own Gemini. As users increasingly turn to LLMs for web searches, Google responded by placing AI-generated summaries at the top of its search results.

The way people navigate the internet is quickly evolving, with one trend reshaping the business models of online companies: the zero-click search. According to a Bain & Company survey, consumers now default to accepting AI-generated answers without further interaction. The data is striking: 80% of users report being satisfied with AI responses for at least 40% of their searches, often stopping at the summary page.

Threat to ad revenue

This AI-driven shift in consumer behavior threatens not only Google’s business model but also that of most internet-based companies. Advertising accounts for roughly 80% of Google’s revenue, earned by charging companies for prominent placement in search results and by leveraging its vast amount of user data to sell ad space across the web. If users stop clicking links, this revenue stream evaporates.

More importantly for this ruling, the market Google once monopolized may no longer be the relevant one. Today, Google’s primary potential competitors in search are not Microsoft Bing, but AI models like ChatGPT, Claude, and Perplexity. In the global race for AI dominance, the outcome is far from certain.

From an antitrust standpoint, there is little justification for penalizing Google now or forcing it to cede advantages to competitors. What would be the benefit for consumers of forcing Google to accept the £24.6 billion offer from Jeff Bezos’ Perplexity AI to buy the Chrome browser?

In essence, the judge acknowledges that Google monopolized the search engine market for a decade but concludes that the issue may resolve itself in the years ahead.

This situation echoes the first major monopolization case: Internet Explorer. For years, European and US regulators battled Microsoft to dismantle the dominance of its web browser, which was bundled with the then-dominant Windows 95 operating system.

By the time all appeals were exhausted, however, the monopoly had vanished. Internet Explorer was partly a victim of the rise of smartphones, which did not rely on Windows. The new king in town was a newcomer: a certain Google Chrome.

How you view the economic and political power of tech giants will shape which lesson you draw from this story. An optimistic view I suggested (with the economist Jana Friedrichsen) is that winner-takes-all markets can intensify competition through innovation. In such markets, incremental investment is not enough; to challenge Google, a competitor must offer a vastly superior product to capture the entire market.

Precisely because they ruthlessly defend their monopoly positions, tech giants show competitors that the potential gains from radical innovations are massive. The pessimistic view, however, is that years of dominance have left these firms largely unaccountable, which could embolden them in future.

This article is republished from The Conversation under a Creative Commons license. Read the original article.

Competition is seen as a panacea in electricity markets: if only we had more, prices would be lower, and investment and supply security would be higher.

Politicians love this story because it offers respite when electricity prices rise. Just unleash regulators and competition authorities to “fix” competition barriers—problem solved (for now).

Encouraging retail competition becomes a priority. Consumers are slow to change retailers, even if they could save hundreds of dollars a year, which is seen as a brake on competition.

Regulators and policymakers therefore champion price comparison services and other measures to encourage electricity customers to shop around.

Also, standalone retailers often protest that they can’t access generation from their rival “gentailers”—firms that combine electricity generation and retailing—on fair terms.

If only they could—and customers more keenly switched providers—retail-only companies could provide stiffer competition. Their solutions include lobbying for gentailers to be broken up, or be forced to supply retailers on the same terms as the gentailers’ own retail arms.

The trouble is, if we misidentify the causes of lackluster electricity market competition, our solutions may only make things worse.

Rather than the lack of competition being about too little customer switching and barriers to retailers entering the market, the more likely cause is too much of both.

Hit-and-run retailers

For the big gentailers (such as New Zealand’s Mercury, Meridian, Contact and Genesis) to face more competition, we need either more gentailers or other ways to achieve the benefits of gentailing. Those benefits are twofold:

• combining generation with retailing effectively manages the huge risks standalone generators or retailers face when they buy and sell on wholesale markets, where prices are highly volatile and can rise to levels that kill businesses; in turn, this helps gentailers finance investment in generation
• and gentailers only need to add one profit margin to their generation cost when setting retail prices; separated generators and retailers add separate margins, which can accumulate to more than what gentailers alone charge.

Separating generation from retailing is therefore a bad idea—if you want lower prices and better investment, you probably want more gentailing.

But why can’t separated generators and retailers replicate these gentailing advantages through long-term contracts? Because generators incur large investment costs to be recovered over many years, so to finance their investments they need long-term revenue security.

Standalone retailers can’t credibly sign contracts offering that security. If they do, new retailers (which can be set up relatively cheaply) can steal their customers when wholesale prices fall below the level of those long-term contracts.

If retailers do sign long-term contracts with generators, they risk failing when exposed to such “hit-and-run” competition by rival retailers—or they renege on those contracts to survive.

Generation investors see this coming, so don’t contract long-term with standalone retailers. Result: lack of viable investment and competition by separated generators and retailers.

The right kind of competition

To resolve this, we would need to eliminate hit-and-run retail entry—first, by making it harder for customers to change retailers if wholesale prices fall below long-term contracted prices.

This could be achieved by requiring retail customers to sign up to long-term retail contracts themselves, rather than being able to flexibly change retailers. Ironically, price comparison websites take us in the wrong direction.

Second, new retailers could be required to have either their own generation—be gentailers, in other words—or have long-term supply contracts in place with generators.

Counterintuitively, this actually makes it easier—or at least more sustainable—for retailers to enter the market, because they know they won’t face hit-and-run competition if they do.

This also means generators can more confidently sign long-term contracts with retailers. Retailers wouldn’t then need to convince regulators to force gentailers to supply them, as they can secure their own supply through contracting.

Standalone retailers might object that they would do this now if they could. But generators can’t supply standalone retailers given the current long-term contracting uncertainty.

Fix that uncertainty—by increasing the ability of retailers to commit to long-term contracts—and both generators and retailers win. Ultimately, this means gentailers face more credible competition, which also means consumers win.

By discouraging the wrong kind of competition (rather than promoting it), genuine competition can be made more durable and effective. That would support long-term investments by generators, and also investments by retailers in innovative services that benefit consumers.

Neither is possible when customers can change retailers with ease, and retailers face hit-and-run competition. If we want more competitive electricity markets, we need to encourage the right type of competition—by discouraging the wrong type.

This article is republished from The Conversation under a Creative Commons license. Read the original article.

Anthropic will pay at least $1.5 billion to settle a US class action lawsuit over allegedly using pirated books to train its artificial intelligence models, according to court documents filed Friday.

“This landmark settlement far surpasses any other known copyright recovery,” said plaintiffs’ attorney Justin Nelson. “It is the first of its kind in the AI era.”

The settlement stems from a class-action lawsuit filed by authors Andrea Bartz, Charles Graeber, and Kirk Wallace Johnson, who accused Anthropic of illegally copying their books to train Claude, the company’s AI chatbot that rivals ChatGPT.

In a partial victory for Anthropic, US District Court Judge William Alsup ruled in June that the company’s training of its Claude AI models with books—whether bought or pirated—so transformed the works that it constituted “fair use” under the law.

“The technology at issue was among the most transformative many of us will see in our lifetimes,” Alsup wrote in his decision, comparing AI training to how humans learn by reading books.

However, Alsup rejected Anthropic’s bid for blanket protection, ruling that the company’s practice of downloading millions of pirated books to build a permanent digital library was not justified by fair use protections.

“We remain committed to developing safe AI systems that help people and organizations extend their capabilities, advance scientific discovery, and solve complex problems,” Anthropic deputy general counsel Aparna Sridhar said in response to an AFP inquiry.

San Francisco-based Anthropic announced this week that it raised $13 billion in a funding round valuing the AI startup at $183 billion.

Anthropic competes with generative artificial intelligence offerings from Google, OpenAI, Meta, and Microsoft in a race that is expected to attract hundreds of billions of dollars in investment over the next few years.

Thousands of books

According to the legal filing, the settlement covers approximately 500,000 books, translating to roughly $3,000 per work—four times the minimum statutory damages under US copyright law.

Under the agreement, Anthropic will destroy the original pirated files and any copies made, though the company retains rights to books it legally purchased and scanned.

“This settlement sends a strong message to the AI industry that there are serious consequences when they pirate authors’ works to train their AI, robbing those least able to afford it,” said Mary Rasenberger, CEO of the Authors Guild, in a statement supporting the deal.

The settlement, which requires judicial approval, comes as AI companies face growing legal pressure over their training practices.

A US judge in June handed Meta a victory over authors who accused the tech giant of violating copyright law by training Llama AI on their creations without permission.

District Court Judge Vince Chhabria in San Francisco ruled that Meta’s use of the works to train its AI model was “transformative” enough to constitute “fair use” under copyright law.

Apple Intelligence

Meanwhile, Apple on Friday was targeted with a lawsuit by a pair of US authors accusing the iPhone maker of using pirated books to train generative AI built into its lineup of devices.

The tech titan’s suite of capabilities called “Apple Intelligence” is part of a move to show it is not being left behind in the AI race.

“To train the generative-AI models that are part of Apple Intelligence, Apple first amassed an enormous library of data,” read the suit.

“Part of Apple’s data library includes copyrighted works—including books created by plaintiffs—that were copied without author consent, credit, or compensation.”

Apple “scraped” works from sources including “shadow libraries” stocked with pirated books, the suit contends.

Apple did not immediately reply to a request for comment.

The suit filed against Apple by Grady Hendrix, author of “My Best Friend’s Exorcism,” and Jennifer Roberson of Arizon, whose books include “Sword-Bound,” seeks class action status.

© 2025 AFP