Early in the new year, when millions of people logged into Google Mail, they were confronted with a tool they hadn’t signed up for. Gemini, Google’s artificially intelligent assistant, was summarizing their emails, and there was no practical way to turn it off.
The shift felt similar to what happened two years ago when the company started showing AI Overviews, or automatically generated responses to people’s questions, at the top of Google search results, with no ability to opt out.
Google’s tactics mirrored Meta’s deployment of its own A.I. chatbot, Meta AI, which became an unremovable tool inside apps including Instagram, WhatsApp and Messenger.
The impact of this A.I. takeover has been subtle but significant. The internet is beginning to look different for everyone, with tailored ads, bespoke advice and unique product prices shown to people depending on what they say to the chatbots. And there is typically no “off” switch.
To put it another way, the tech industry is making a personalized internet just for you, but you have no say in it.
“These tools are sold to us as more powerful, but we have less say in things,” said Sasha Luccioni, a researcher who focuses on A.I. ethics at Hugging Face, an A.I. company. “It’s on us to opt out, and it’s usually pretty complicated and not very clear what we should be opting out of.”
The companies have said they are focused on creating the best “agent” — an A.I.-powered assistant that can write emails, book plane tickets and do research — to empower people. Generative A.I. is so flexible, adaptable and capable, they say, the technology can essentially give everyone a unique experience of using the internet through a digital assistant catering to his or her needs.
The tech industry’s strategy of forcing A.I. on the masses is at odds with feedback from many users. Americans are generally more concerned than excited about A.I. in their daily lives, with the majority saying they want more control over how the technology is used, according to a survey last spring by the Pew Research Center.
Google said in a statement that people found A.I.-powered search to be more helpful and were coming back to do more searches as a result. The company added that it offers a “web” tab on Google.com to filter out A.I.-generated results, but that people use it for only a tiny fraction of searches.
Meta said people could choose whether to engage with Meta AI across its apps. However, avoiding interacting with the A.I. would be difficult for most people to do, because the A.I. assistant is part of the search tool in some apps, including Instagram.
The insistence on A.I. everywhere — with little or no option to turn it off — raises an important question about what’s in it for the internet companies. A.I. chatbots, including Gemini and Open AI’s ChatGPT, are enormously expensive to operate and have not directly generated profits for companies through subscription fees, because many people use their free features.
Behind the scenes, the companies are laying the groundwork for a digital advertising economy that could drive the future of the internet. The underlying technology enabling chatbots to write essays and generate pictures for consumers is being used by advertisers to find people to target and automatically tailor ads and discounts to them. Smaller brands and online retailers that fail to adapt could get lost in the A.I.-generated noise.
Last month, OpenAI said it would begin showing ads in the free version of ChatGPT, based on what people were asking the chatbot and what they had looked for in the past.
In response, a Google executive mocked OpenAI, adding that Google had no plans to show ads inside its Gemini chatbot. What he didn’t mention, however, was that Google, whose profits are largely derived from online ads, shows advertising on Google.com based on user interactions with the A.I. chatbot built into its search engine.
(The New York Times has sued OpenAI and Microsoft, claiming copyright infringement of news content related to A.I. systems. The two companies have denied the suit’s claims.)
For the last six years, as regulators have cracked down on data privacy, the tech giants and online ad industry have moved away from tracking people’s activities across mobile apps and websites to determine what ads to show them. Companies including Meta and Google had to come up with methods to target people with relevant ads without sharing users’ personal data with third-party marketers.
When ChatGPT and other A.I. chatbots emerged about four years ago, the companies saw an opportunity: The conversational interface of a chatty companion encouraged users to voluntarily share data about themselves, such as their hobbies, health conditions and products they were shopping for.
The strategy already appears to be working. Web search queries are up industrywide, including for Google and Bing, which have been incorporating A.I. chatbots into their search tools. That’s in large part because people prod chatbot-powered search engines with more questions and follow-up requests, revealing their intentions and interests much more explicitly than when they typed a few keywords for a traditional search.
Because chatbots are helping companies collect more detailed, intimate information about people’s interests, the newer forms of digital advertising can feel more invasive and uncanny, experts say.
Based on what a person has been saying to a chatbot, among other data sources, both Meta’s and Google’s newer advertising systems could potentially infer, for instance, that the person likes to run in the winter and then use A.I. to automatically create an ad on Instagram or Google with marketing language about the durability of a pair of shoes in cold weather.
For brands, the key is to personalize an ad enough without making it noticeable that A.I. was involved, said Luke Stillman, the managing director of Madison and Wall, a consulting firm for online merchants that do advertising.
“I think of all this A.I. optimization as bad plastic surgery,” Mr. Stillman said. “You notice the ones that stand out, but you don’t notice the ones that are still standing behind the curtain.”
The information gleaned from chats with Gemini and other data could also eventually affect the prices different people see for the same products. Last month, Google unveiled an A.I.-powered shopping tool that it developed with retail companies including Shopify, Target and Walmart. The new technology can help merchants automatically base their product prices on information that consumers have shared with the Gemini chatbot, such as their personal budget, among other data sources.
A clothing brand, for example, could offer a $32 discount on a $120 jacket to a more frugal shopper but offer the same jacket at full price to someone with a higher income.
Lindsay Owens, executive director of the Groundwork Collaborative, a nonprofit that focuses on economic issues, called Google’s A.I. shopping framework an example of “surveillance capitalism” that could eventually be used to goad people into spending more.
“Knowing my spouse’s birthday is tomorrow is of interest to a retailer because they can charge me more for flowers,” Dr. Owens said. “This could change shopping as we know it.”
Smaller rivals to Google and Meta, including Mozilla, the maker of the Firefox web browser, and DuckDuckGo, a search engine with a focus on data privacy, have called out the lack of control that users have over how they use A.I.
This month, Mozilla said a new version of Firefox would include a broad set of controls to turn on or off individual A.I. features. Anthony Enzor-DeMeo, Mozilla’s chief executive, said the risk of putting A.I. everywhere was making the web less open, because only the most privileged users can pay subscription fees to take advantage of the most powerful models.
“A.I. is changing how people browse the internet, so Firefox and Mozilla need to change with it, but that doesn’t mean we have to go about it in a way that’s going to alienate or piss people off,” he said.
Last month, DuckDuckGo launched a version of its search engine that filtered out any results containing images generated with A.I. The company asked users whether they did or did not want to use A.I. About 90 percent of respondents voted “No A.I.”
But even if smaller companies are offering methods to turn off A.I. in their products, it would still be difficult to avoid the A.I. baked into products made by Google and Meta, whose tentacles reach into nearly everyone’s life through services like email, word processing, text messaging and social networking apps. Mark Zuckerberg, Meta’s chief executive, said in the company’s recent earnings call that 3.58 billion people, or roughly 44 percent of the world, use at least one Meta product per day. Google’s share of the global search market has held steady at around 90 percent.
Kamyl Bazbaz, an executive at DuckDuckGo overseeing public affairs, marveled at Google’s power to keep people loyal.
“Big companies that have users locked in can get it wrong for a very long time before they ever get fired by a user,” Mr. Bazbaz said. “AI Overviews on Google, everyone thought it was bad for at least two years. And it’s not like people left.”
Brian X. Chen is the lead consumer technology writer for The Times. He reviews products and writes Tech Fix, a column about the social implications of the tech we use.
The post A.I. Is Giving You a Personalized Internet, but You Have No Say in It appeared first on New York Times.
