Analysis | Her guacamole recipe tops Google. She fears AI will change that. (2024)

If you Google “guacamole,” there’s a good chance your top result will be Lisa Bryan’s recipe. Titled “Best Ever Guacamole (Fresh, Easy & Authentic),” it calls for a classic mélange of avocados, Roma tomatoes, cilantro, garlic, onion, lime, jalapeño and sea salt.

That prime placement on a popular search query is Bryan’s meal ticket. But she fears artificial intelligence will soon snatch it away.

A former health-care executive from Southern California, Bryan burned out in her career a decade ago and started posting recipes online for family and friends. Now she runs a food and lifestyle blog called Downshiftology, where she advocates “taking life down a notch” and savoring simple pleasures. She employs a full-time social media manager, has 2.5 million YouTube followers and says her website reaches 130 million people a year.

Hers is a success story made possible in large part by Google Search, which directs millions of people to her blog — with noticeable boosts ahead of the Super Bowl and Cinco de Mayo, when searches for guacamole peak. But as Google shifts from traditional search results toward answering users’ questions directly with AI, independent web publishers like Bryan fear for their livelihoods.

Now the bloggers are taking their case to Congress.

On Wednesday, they’re staging an “Independents’ Day” lobbying push on Capitol Hill. The push is being organized by a company called Raptive, which handles advertising and marketing for online publishers and helps them rank highly in search results — giving it a vested interest in beating back AI.

Bryan is among thousands who signed onto an open letter to Congress from Raptive CEO Michael Sanchez urging scrutiny of Google’s “AI Overviews.” Several of those creators will also meet with staffers and lawmakers from their home states.

“This new product takes revenue and copyrighted content from publishers and creators without consent or compensation, competing directly with creators and giving nothing in return,” the letter argues. It calls on Congress to push tech companies to pay content creators when their work is used to train AI tools, and to press Google to commit that its AI answers won’t reduce traffic to third-party websites.

Raptive shared the open letter and an accompanying report — including survey results from its clients about the importance of Google search to their business — with the Tech Brief.

The group isn’t pushing specific legislation but will ask lawmakers hold hearings and pressure the tech industry to do right by web creators.

While big media companies may be able to bring tech giants to the negotiating table, Sanchez said, Raptive’s clients don’t have that kind of clout. Among the creators meeting with lawmakers are a crochet and craft designer from Michigan, a personal blogger from Minnesota, and two brothers from Tennessee who run the country music site Country Rebel.

Bryan didn’t make the trip to D.C. but said she has been urging her audience to trust and follow individual creators over AI answers. “All of my recipes are tasted and tested, and that is something a chatbot cannot do,” she said.

Google contends that the fear of AI answers disrupting the web economy is misplaced.

The feature is intended to complement search results, not replace them, Google spokesperson Brianna Duff said. AI Overviews only show up on certain searches, when Google’s systems predict that they can be especially helpful — such as when the user wants a summary of information from a range of sources.

The company has also said sources cited in an AI Overview get more traffic than those that appear in traditional search results. And CEO Sundar Pichai told the Verge in May that Google recognizes the value of the web ecosystem, which its AI answers depend on.

Still, the feature stumbled out of the gate, as users posted viral screenshots of Google’s AI suggesting that people eat rocks or put glue in their pizza. In response, the company has temporarily pulled back on the number of searches that trigger AI responses while it tries to work out the kinks. But some experts say the problem of low-quality answers may not be fully solved anytime soon.

Web creators are the latest group to join a growing backlash against tech firms’ unauthorized use of their work to train chatbots, image generators and other generative AI tools.

Artists, authors and media organizations have sued OpenAI, the maker of ChatGPT, and other AI companies alleging copyright infringement. Others have struck licensing deals allowing the firms to use their work for a price. This week, a group of record labels sued a pair of AI music companies that use software to generate songs on-demand based on a user’s prompts.

How the lawsuits will play out remains an open question. Tech firms argue that training AI systems on others’ published work without permission amounts to “fair use” under copyright law, because their software transforms the work into something new and original.

But they seem to be losing ground in the court of public opinion, and some tech executives’ public statements aren’t helping.

OpenAI’s chief technology officer, Mira Murati, drew criticism this week for a recent talk in which she said that “some creative jobs maybe will go away” as AI tools replace them while assuring that those tools will help others be more creative. As for the jobs lost in the bargain, she added, “maybe they shouldn’t have been there in the first place.”

Bret Taylor, Sierra co-founder and CEO and chair of OpenAI, spoke with The Post’s Elizabeth Dwoskin about the present and future of AI at the Post Live event The Futurist: The New Age of AI.

That’s all for today — thank you so much for joining us! Make sure to tell others to subscribe to Tech Brief. Get in touch with Cristiano (via email or social media) and Will (via email or social media) for tips, feedback or greetings!