Hello, and welcome to TechScape. I’m your host, Blake Montgomery, writing to you as I finish the audiobook version of Don DeLillo’s White Noise, which I can’t say I found compelling.
In tech – artificial intelligence is having its day in court with an 11th-hour appearance in Google’s landmark antitrust trial and Anthropic’s major settlement with book authors.
Why OpenAI helped Google skirt a Chrome sale
Google dodged a catastrophic breakup, and it has its biggest competitor to thank for that, according to the judge who could have forced the tech giant to sell off Chrome, the most popular web browser in the world, and perhaps Android, the world’s most widely used mobile operating system.
Amit Mehta, who ruled in 2024 that Google had built and maintained an illegal monopoly over the internet search business, said last week that he would not force the most drastic remedy on the tech giant. The US government had requested that Mehta enjoin Google to divest Chrome. Instead, the company is prohibited from entering into exclusive distribution agreements for its search engine – but not barred from distribution agreements entirely – and must share data from its search business with competitors. Google is expected to appeal, but for now, one can imagine Sundar Pichai breathing a sigh of relief.
Critics said the penalties were weak, nothing more than a “slap on the wrist”, a phrase that appeared at least half a dozen times in the statements that flooded my inbox after the ruling came down.
The reason for the relative tameness of the penalty is the emergence of real competition to Google – what the case concerned in the first place. United States v Google may have focused on search, but its aftermath concerned generative artificial intelligence.
“The emergence of GenAI changed the course of this case,” Mehta wrote. “These remedies proceedings thus have been as much about promoting competition among general search engines as ensuring that Google’s dominance in search does not carry over into the generative AI space.”
Mehta said that in a prior era, internet search attracted neither investment nor innovation, so great was the fear of and dominance exerted by Google. That is no longer the case. Generative AI companies are attracting hundreds of billions of dollars in investment to create new products that threaten the primacy of internet search. It is no secret which company Mehta is referring to: he namechecks OpenAI and ChatGPT 30 times each in his ruling.
“These companies already are in a better position, both financially and technologically, to compete with Google than any traditional search company has been in decades,” Mehta wrote. “These new realities give the court hope that Google will not simply outbid competitors for distribution if superior products emerge. It also weighs in favor of ‘caution’ before disadvantaging Google in this highly competitive space.”
Google has been Safari’s default search option since the iPhone debuted nearly 20 years ago. By contrast, you can see the competition in generative AI in action in Apple’s treatment of Google and OpenAI. In June 2024, Apple announced that it would partner with OpenAI on features for the iPhone. In August 2025, though, Apple engaged in talks with Google about using Gemini to revamp the beleaguered Siri, according to Bloomberg. May the best bot win.
In April, I predicted that OpenAI would emerge as a potential buyer for Chrome, would perhaps even take a chance on buying it. At the very least, the ChatGPT maker would benefit from a weakened Google. Later that month, an OpenAI executive testified that the company would want to do just that.
It is poetic that OpenAI’s success saved Google. Perhaps the upstart owes the favor to its forebear. After all, a paper produced by Google researchers did provide the theoretical basis for ChatGPT. Researchers from the Mountain View campus proposed the foundational architecture that would make ChatGPT possible in 2017.
Google, valued at $2.84tn, is surely the Goliath in this scenario to the David of OpenAI, possibly worth $500bn. Lest you think OpenAI is not Google’s biggest competition, put your ear close to the horse’s mouth. In December 2022, Google’s own management declared the nascent popularity of ChatGPT a “code red” for its lucrative search business, which brings in hundreds of billions of dollars a year. Pichai himself reassigned vast hordes of Google’s employees to artificial intelligence projects.
Unlike Goliath, who disregarded his puny, slinging challenger, Google itself acknowledged that ChatGPT’s release, which marked the arrival of generative AI in the popular consciousness, had changed the competitive landscape. The threat was grave.
Google has caught up to OpenAI in the AI race, but the David in the fight still holds the sling of first mover advantage. ChatGPT remains synonymous with generative AI, and perhaps AI in general, in the minds of many people. Google is enormous, though, and still the incumbent player. Billions of people encountering its search engine’s AI Overviews on a daily basis.
Google may have dodged catastrophe with Mehta’s decision to allow Chrome to remain in the fold. The wolves, however, are still circling. The tech giant faces another antitrust hearing later this year regarding the technology that underpins its advertising business, the financial engine of the company. Google owns the distribution mechanisms of online ad sales, and the digital location where the sale is taking place.
The European Union fined Google nearly €3bn for abusing its dominant market position in advertising technology the same week as Mehta’s decision came down and has threatened a breakup of the company’s adtech division.
Read more
after newsletter promotion
Tech in the UK
A big payout gives authors hope to cash in on AI
Last week, Anthropic, maker of the Claude chatbot and one of the most prominent startups in the generative artificial intelligence realm, agreed to pay a group of authors $1.5bn to settle a lawsuit accusing it of pirating millions of books to train its AI. The plaintiffs called it the largest copyright recovery ever. Anthropic admitted no liability but offered about 500,000 authors $3,000 per work for the total payout of $1.5bn.
The company had acknowledged using major databases of pirated books – Books3, Library Genesis, and Pirate Library Mirror – to train its AI models, in total approximately 7 million books downloaded in 2021. As the threat of copyright litigation loomed in 2024, it acquired and scanned millions of physical copies of those same, then destroyed them. What a trail of waste.
The settlement is a victory for the creative professionals who say that AI poses an existential threat to their livelihoods, both from the permissionless use of their works to create generative software and the flood of AI slop that captures users’ attention over human-created works. Authors were among the first to sound the alarm, as AI could produce text first. Writers in the UK have called for their government to hold Meta accountable for what they see as theft. In the US, where the CEO of Meta is sitting side-by-side with the president at dinner in the White House, any action hostile to tech companies’ growth seems unlikely.
The news of Anthropic’s settlement has already inspired others. The same day the news broke, authors filed suit against Apple on the grounds it similarly used Books3 to train its AI.
Anthropic’s settlement is not an unqualified victory for the writers, though. The legal problem for the AI company was more the piracy, a much more blatantly illegal act with established precedent for penalties. The judge in the case had ruled in July that the meat of the case, the thing authors seemed more upset about – Anthropic’s use of copyrighted books as training data for its AI – was permissible under fair use, a legal doctrine that allows for the reproduction of copyrighted material with sufficient transformation. Judge William Alsup compared the Anthropic model’s use of books to a “reader aspiring to be a writer”. It seemed the AI company would win the case outright just two months ago.
Read more: Anthropic did not breach copyright when training AI on books without permission, court rules
Meta is the likeliest next target of authors interested in litigation, for my money. Like Anthropic, the company has admitted using the pirated databases Books3 and Libgen to train its Llama models. Meta won a favorable victory similar to Anthropic’s in its own copyright fight in July, but the startup’s settlement seems likely to motivate Meta’s lawyers to shut down the litigation as soon as possible.
The other major AI players stand farther away from the authors gunning at them. OpenAI and Microsoft have been accused of using Books3 in lawsuits, but that usage has not been confirmed in the same way as with Anthropic or Meta.
A ream of other copyright suits over AI are in motion, though they apply to other media. Warner Bros. Discovery sued Midjourney last week; Disney and Universal did the same in June.
