More than half of your potential customers are getting purchase advice from a chatbot. And that chatbot is telling them not to buy.
That's not a dystopian prediction. That's what's happening right now. A survey of over 2,300 U.S. consumers by Exploding Topics found that 57.5% of AI users have decided against a purchase based on information from a chatbot. Among weekly AI users, that number climbs to nearly 70%.
Let that sink in for a second. We've spent years optimizing funnels, A/B testing landing pages, and obsessing over attribution models. Meanwhile, a conversational AI is quietly intercepting customers at the research stage and steering them away from the checkout button.
Marketing is like dating, I always say. You don't propose on the first ad impression. But what happens when a third party crashes your date and whispers in your prospect's ear that maybe you're not the one?
The New Gatekeeper Doesn't Work for You
Here's the thing about digital marketing in 2026: the funnel has a new bouncer, and you didn't hire them.
G2's research from earlier this year shows that 51% of B2B software buyers now start their purchasing process in an AI chatbot rather than a traditional search engine. Not Google. Not your website. ChatGPT, Claude, Perplexity. That's where the journey begins.
And here's where it gets interesting: 69% of those buyers ended up choosing a different vendor than they initially planned based on AI chatbot guidance. A third purchased from a vendor they'd never heard of before.
The Yellow Pages compressed the market into the big book. Google compressed it into the first page of results. Now, AI chatbots are compressing it into a single answer.
Tim Sanders, G2's Chief Innovation Officer
One answer. Not ten blue links. Not a sponsored carousel. One synthesized recommendation that either includes your brand or doesn't.
The Trust Paradox
Now, before you panic and redirect your entire budget to "AI optimization" (whatever that means this week), let's talk about the trust problem.
Akeneo's consumer research found that only 45% of consumers actually trust AI-powered recommendations. And academic research published in The Conversation shows 71% of customers still prefer interacting with a human agent over a chatbot.
So consumers don't trust AI recommendations, but they're still using them to make purchase decisions? That's not a contradiction. That's human behavior in all its messy glory.
Think about it this way: I don't fully trust Yelp reviews either. I know some are fake, some are from people with wildly different taste than mine, and some are revenge posts from disgruntled ex-employees. But I still check Yelp before trying a new restaurant. The tool doesn't need to be perfect to be influential.
The same dynamic is playing out with AI chatbots. Consumers are skeptical of the technology in the abstract while simultaneously letting it shape their purchasing behavior in practice. They're using chatbots as a filter, a sanity check, a way to feel more confident about decisions they were already leaning toward.
The Organic Recommendation Gold Rush
Here's where the opportunity lives. The same Exploding Topics survey found that 41% of consumers have made a purchase based explicitly on an organic AI recommendation. Among weekly AI users, 73.58% have bought something because a chatbot suggested it.
That's not paid placement. That's not sponsored content. That's the AI synthesizing information about your brand from across the web and deciding you're worth recommending.
The question every CMO should be asking right now: What is ChatGPT saying about my brand when someone asks for a recommendation in my category?
If you don't know the answer, you're flying blind. And if the answer is "nothing" or "something negative," you've got a problem that no amount of paid media can solve.
Why Chatbot Ads Aren't the Easy Fix
The obvious response to "AI is intercepting my customers" is "let's buy ads in AI." And sure, that's an option. But the data suggests it's not a silver bullet.
More than 4 in 10 people actively dislike AI chatbot ads, and the average consumer's opinion of a brand is slightly more likely to worsen than improve after seeing an ad in a chatbot. Among all AI users, 71.63% feel neutral or negative toward ads in AI chatbots.
This tracks with what we know about interruptive advertising in general. People don't go to ChatGPT to see ads any more than they went to Google to see ads. They tolerate them. Sometimes they click. But the context matters enormously.
When someone asks an AI chatbot "what's the best project management software for a 50-person marketing team," they're in research mode. They want information, not interruption. An organic mention in the chatbot's response carries weight precisely because it feels like advice rather than advertising.
The GEO Imperative
So what do you actually do about this?
First, accept that Generative Engine Optimization (GEO) is no longer optional. It's not a nice-to-have for forward-thinking brands. It's table stakes for not losing customers you've already spent money to attract.
Second, audit your AI presence. Ask the major chatbots about your category. Ask them to compare you to competitors. Ask them what they'd recommend to someone with your ideal customer profile. The answers might surprise you, and not in a good way.
Third, recognize that the inputs to AI recommendations are the same things that have always mattered: genuine customer reviews, authoritative content, consistent brand messaging across channels, and actual product quality that people talk about. The chatbot is synthesizing signals that already exist. If those signals are weak or negative, no amount of prompt engineering will save you.
Fourth, don't abandon your existing channels. Almost two-thirds of AI users have only partially replaced Google searches with AI conversations. The old playbook still matters. It just has company now.
The Real Question
Data tells you the what, but brand tells you the why. The data here is clear: AI chatbots are influencing purchase decisions at scale, and that influence cuts both ways. They're creating customers and killing deals.
The why is more interesting. Consumers are overwhelmed. They're drowning in options, reviews, comparison sites, and marketing messages. AI chatbots offer something seductive: a shortcut. One answer instead of ten tabs. A recommendation instead of a research project.
That's not going away. If anything, it's accelerating. Over 987 million people worldwide now use AI chatbots, and the market is projected to double in the next few years.
The brands that win in this environment won't be the ones who figure out how to game the algorithm. They'll be the ones who build products and experiences worth recommending, then make sure the AI knows about it.
Marketing has always been a team sport. Now there's a new player on the field, and it's got opinions about your brand whether you like it or not.
