Introduction: How AI Changes Buyer Vetting
I’ve been writing about what I call the Trust Deficit Economy: the economic environment created as cheap capital disappeared and businesses shifted from prioritizing growth to prioritizing profitability. Another place this shift is becoming increasingly visible is how buyers gather information and who they trust to provide it.
The growing buyer skepticism of salespeople predates AI. What AI changed is the speed and cost of verification. Buyers can now research products, validate features, investigate implementation requirements, compare claims, and arrive at a sales conversation with substantially more information than they could gather on their own before.
This article explores what happens to sales when the salesperson is no longer the unquestioned source of truth about the thing they sell.
Buyers did not get sharper because of AI
AI did not create buyer skepticism. It dramatically lowered the cost and time required to act on it.
That skepticism has been building for decades. The internet, review platforms, forums, peer networks, and social media progressively erased the information advantage sellers used to hold. Buyers then learned that “independent” information could be manipulated too. Paid placements, manufactured reviews, sponsored content, SEO, and backlink campaigns could all be built to look like organic authority.
So buyers learned to triangulate. Cross-check one channel against another.
The research supports how significant this trust problem has become.
Forrester's 2025 research found coworkers and management were trusted by 82% of B2B buyers, current vendors by 79%, and independent experts such as peers and analysts by 66% to 72%.
Salespeople from prospective vendors landed among the least-trusted information sources.
The trust deficit was not created by AI.
A few years back, a client was deeply within a persuasive sale gone bad. They had purchased an LMS platform along with six figures of professional services. Once delivery started, the expertise the sales process had promised was not showing up in the work.
We started asking questions.
Who are the professionals actually doing this work? What are their backgrounds? What credentials do they hold? What certifications and professional training do they have?
The answers were not good.
- The “graphic design” services were effectively being delivered by people trained to use PowerPoint without professional graphic design backgrounds.
- The instructional design work was coming from the same delivery pool rather than credentialed learning and development professionals.
The company had paid premium professional-services pricing for expertise that did not meaningfully exist in the form it was sold.
Here is where the story gets more interesting: I don't think the salesperson knew.
I obviously cannot prove what they knew internally. But watching them respond as we raised the problems, struggle to answer basic questions about credentials, and repeatedly pull leadership into the conversation, my read was that they were discovering the delivery problem alongside us.
Leadership knew how its professional-services organization was staffed. I am far less convinced the individual seller did.
That distinction changes what the trust problem actually is: leadership decisions.
Salespeople can be unreliable sources of truth while acting completely in good faith.
A salesperson can accurately repeat the positioning, product claims, ROI numbers, implementation expectations, competitive talking points, and service descriptions their company taught them. None of that guarantees those claims are real.
It took us four months to negotiate a refund for services that did not match what had been represented. We eventually recovered the money and hired properly qualified professionals for roughly half of what the original vendor had charged.
That experience teaches buyers that the salesperson may not know whether what they are telling you is true. This means that even if the buyer builds a relationship with the seller, the trust deficit can still remain.
AI doesn't solve the trust problem
Buyers are not moving toward AI because they suddenly found an information source they can trust without question. AI hallucinates. It repeats bad information. It can surface manipulated information just as confidently as accurate information.
Gartner's 2026 research captures this almost perfectly. When buyers were asked where they were more likely to encounter misleading information, 51% said GenAI and 49% said a sales representative.
That is essentially a tie.
Yet those same buyers are using both and comparing notes.
Gartner found 67% of B2B buyers prefer a rep-free experience, 70% prefer completely digital self-service, and 45% used GenAI during a recent purchase.
Buyers reported using an average of seven information sources during a purchase.
At the same time, 69% said they turn to sales representatives specifically to validate AI-generated insights.
Those numbers look contradictory if we assume buyers are trying to decide whether they trust AI or the salesperson.
I don't think they are.
I think they are checking one source against another.
AI dramatically reduced the difficulty verification
Imagine trying to perform serious independent due diligence on an enterprise software purchase fifteen years ago.
You could search Google. You could talk to peers. You could read whatever reviews existed. You could request references. You could research competitors individually. You could dig through documentation and try to reconcile what you found against what the salesperson told you.
But it took time.
AI compresses that work.
A buyer evaluating the LMS provider I described earlier could now ask what credentials legitimate instructional designers typically hold, what professional graphic design services should include, what comparable firms charge, what customers repeatedly complain about, what implementation problems appear across reviews, which claimed features are actually native to the product, what competitors offer, and what questions they should ask the seller before signing.
That does not guarantee correct answers.
What it provides is an extraordinarily fast first pass at where the buyer should look harder.
Forrester's 2026 buying research shows just how quickly this behavior is becoming embedded. 94% of business buyers reported using AI during the buying process. Buyers increasingly use AI as a starting point, but then validate what it produces against peers, product experts, analysts, colleagues, and other sources they trust.
This isn't the replacement of human due diligence. It's the acceleration of it.
The salesperson has moved from information provider to information validator
For most of modern sales history, product knowledge gave the salesperson an information advantage. The salesperson knew the product. They knew the category. They knew the implementation. They knew what competitors supposedly could and could not do.
The buyer needed access to that knowledge.
That information advantage has been deteriorating for decades. AI accelerates the final stage of that erosion because it does more than give buyers access to information. It helps them synthesize it.
Forrester found 73% of B2B purchases now involve three or more departments, with an average of 13 people inside the organization and nine outside influencers involved in the decision. Buyers are consulting colleagues, communities, peers, analysts, customers, AI tools, and vendors across that network.
Gartner reaches an important conclusion from its own data: sellers are shifting from being the primary source of information toward becoming a source of validation and confidence at critical points in the decision.
I think that change is bigger than it sounds.
The salesperson no longer gets to establish truth simply because they represent the product.
They increasingly need to prove they deserve to be one of the trusted sources used to establish it.
Product command has a different job now
Sales organizations have spent years debating how much product knowledge sellers really need.
I have heard every version of it.
- “You don't need to know that much about the product.”
- “You just need to know enough to get the next meeting.”
- “That's what the sales engineer is for.”
- “Don't get too technical.”
There was always a weakness in that thinking. In the Trust Deficit Economy, I think it becomes increasingly dangerous.
The purpose of product command is no longer simply being able to give a good demo or answer a feature question. Product command is part of the seller's credibility.
A buyer may arrive having already researched the feature. They may have compared the documentation against three competitors. They may have asked AI where the implementation typically breaks. They may have read customer complaints and discovered a limitation before the first discovery call.
Now they ask the salesperson about it. The quality of that answer becomes a trust test.
A seller who knows the product can explain the nuance, acknowledge the limitation, correct bad information, explain the tradeoff, or bring in the right expert when they genuinely don't know.
A seller without product command guesses. Deflects. Gives the approved talking point. Makes an absolute statement they cannot substantiate. Or confidently repeats something enablement taught them that has not been true for two product releases.
And buyers can now check.
This is why trust cannot be a sales tactic.
Product, marketing, sales, implementation, customer success, and leadership have to operate from the same version of reality. Product claims need to survive implementation. Professional-service descriptions need to match the people actually delivering the work. Competitive claims need to be defensible. Limitations need to be known before a customer discovers them.
Otherwise, buyers, who are comparing notes, will not believe the information they receive.
The sellers who win in the Trust Deficit Economy will not be the ones who sound the most certain.
They will be the ones whose answers keep turning out to be true.