If you’re following the midterm elections, you’re probably paying close attention to James Talarico’s campaign in Texas — one of the key seats Democrats need to win to help flip control of the Senate. These next few weeks are the most important for campaigns like Talarico’s to raise money (and spend it), and more and more people are likely to be googling James Talarico, finding his campaign’s official website, and making a donation. 

Or are they? 

Earlier in the week, campaign tech strategist A.J. Carrillo noticed something unusual: When asked how to donate to James Talarico’s campaign, Google’s AI Overview returned links to three pages – but only one was a link to Talarico’s official campaign website. And, disturbingly, one of the pages Google’s AI suggested looked like it was created to raise money for Talarico but was actually sending donations to Sen. John Hickenlooper’s re-election campaign. This problem isn’t limited to Talarico either. Instead of sending people who were looking to donate to Kansas Senate candidate Adam Hamilton to his official donation page, the AI Overview directed potential donors to unaffiliated PACs using Hamilton’s name to raise for themselves. 

So what’s actually going on here? 

For decades, when you typed something into a search engine, you got back a dozen or so blue links, and it was up to you to choose what to click on. But increasingly, many internet users aren’t searching for links – they’re searching for answers. Whether it’s individuals choosing to use an AI answer engine like ChatGPT or Perplexity, or using the same Google search we all know and love (or not) and getting presented with an AI-generated response instead of links, the result is one single answer based on a handful of sources.

But these AI tools don’t necessarily check who’s behind those sources. They look for pages that seem credible and appear to answer the question that’s being asked. A page that’s named after a candidate, with the candidate’s photo, biography, and a donate button, looks like it matches the question “how do I donate to [candidate’s name]” very well – even if that page was actually created by a scammy super PAC that’s trying to capture interest in a candidate for their own fundraising goals. 

And it’s not just political candidates who are at risk. Lookalike fundraising pages, outdated third-party listings, and scam “relief funds” set up after a humanitarian disaster can all end up in an AI answer, threatening a nonprofit organization’s revenue – and reputation. 

The good news is, even as the search experience evolves, campaigns and charities still have more control over your search results than you might think. 

Three things you can do right now

1. Make your official donation page the easiest answer to find. 

AI tools reward sources that are clear, consistent, and backed up by other trusted sites. Make it easy for them to pick you. Name your pages clearly. Your official donation page title should explicitly say what it is (such as “Donate to James Talarico – Official Campaign Website”) and be linked in your website, social bios, email footers, and press releases. Claim your profiles on aggregators like Ballotpedia, Candid, or Charity Navigator, and keep them up to date. AI tools lean heavily on these sources. 

2. Pay to own your name in search

One thing that Google’s surely not changing anytime soon is that the search engine results page is still a pay-to-play environment. Run brand search ads on your name, and cover phrases that include keywords like “donate” and “contribute” so that your ad shows up at the top of the page. Make sure your ad copy also makes it clear that your website is the official way to donate to your campaign or organization. These clicks are usually the cheapest and highest value you can pay for – and you can’t take for granted that donors are coming to your official donation page without it.

3. Keep checking what AI answer engines and chatbots are saying about you

You can’t fix what you don’t know about, so get in the habit of regularly asking Google, ChatGPT, and other sources, “How do I donate to [insert name]?” or “What is [insert name]’s official website?” Keep a record of what you find, because the answers will likely continue to change as these tools improve and find new sources of information. And don’t be afraid to use the AI tools’ feedback links to report wrong answers, misrepresentations, or scammers.

It doesn’t take a lot to stop AI hijacking

Right now, some of the money meant for you could be going somewhere else, and an AI tool is helping send it there. But it doesn’t take much time or budget to protect your most valuable asset: your name. Pages that are easy to find, a modest investment in branded search, and getting in the habit of checking what AI answer engines are saying about you can go a long way.