Here’s today’s AdExchanger.com news round-up… Want it by email? Sign up here.
Trust Issues
Google Search is officially a monopoly.
US District Court Judge Amit Mehta ruled on Monday that Google has been operating an illegal monopoly over online search for years, Politico reports.
In a 286-page ruling, Mehta says Google is a monopolist in part because it maintains 90% of the search market through exclusive search provider agreements with mobile carriers and device makers, including Apple, Samsung and Verizon. Google paid billions per year to keep these agreements in effect.
Google also used its dominance of the search advertising market to fend off challenges from Microsoft, one of its chief competitors, Mehta ruled. However, he stopped short of saying Google has a monopoly over search advertising.
It remains unclear what methods the government will pursue to rectify Google’s search monopoly. That will require separate legal proceedings. But Assistant Attorney General for Antitrust Jonathan Kanter has voiced support for “structural remedies” that could include breaking off parts of Google’s business.
Google plans to appeal the ruling.
But the decision will no doubt bolster related efforts to crack down on Big Tech monopolies, since it proves existing antitrust law applies in the internet era. Next up is another Google trial centering on its programmatic ad tech, which is set to begin in September. The US also has pending antitrust cases against Apple, Meta and Amazon.
Scrape, Baby, Scrape
Internal documents leaked to 404 Media reveal that Nvidia has been scraping video content from YouTube, Netflix and other digital sources to train its AI.
Described as “a video data factory that can yield a human lifetime visual experience worth of training data per day” by an Nvidia higher-up, the “Cosmos” project was intended as a foundational AI model for several products, including even self-driving car systems.
Obviously this reopens previous ethical and legal arguments about downloading and using copyrighted content in AI training data sets (YouTube calls this type of scraping a “clear violation” of its terms of use), but it also begs the question: Can we trust AI companies when they say their products are compliant with copyright laws?
As this newsletter previously covered, many advertising agencies are still hesitant to use AI tools without first confirming that their own data will not be used as a training source. More negative press like this won’t exactly inspire confidence in these tools going forward – to say nothing of the many lawsuits now being filed against AI companies, Nvidia included.
Ditto
Warner Bros. Discovery’s flagship service Max just launched a new homepage personalization feature for all US accounts. WBD started rolling it out gradually in December – and it’s awfully similar to the content recommendation setup on Netflix, The Verge reports.
Based on viewing history and consumption patterns, Max now recommends titles in content rows that take over the entire homepage, not just part of it. Max also suppresses titles it doesn’t think a subscriber would be interested in.
Homepage personalization is one tactic streamers are using to increase time spent on their apps, which can raise ad revenue. But another popular strategy for streamers like Max is copying Netflix, which is steadily growing both ad-supported subscribers and ad revenue.
Shortly after Netflix touted all the paid accounts it gained from banning password sharing, for example, Max (and Disney+) announced plans to do the same. And going forward, WBD plans to include more features on Max that encourage user engagement, such as buttons viewers can use to rank the shows they “like” or “love” – which Netflix has been doing for at least two years.
But Wait, There’s More!
Ari Paparo offers a history lesson on how Google built an ads business that’s drawing antitrust scrutiny. [LinkedIn]
The ad job market in New York City is booming – but is that hurting other cities? [Ad Age]
Olympics ads just aren’t hitting like they used to. [The Drum]
The Olympics is increasingly relying on advertising – not media rights – to pay the bills. [Digiday]
What marketers need to know about principal media buying. [WSJ]
Petitioners in Israel drop a class-action lawsuit accusing Bright Data of scraping kids’ data from social media. [Bloomberg]
