Would you let an AI agent spend $50,000 of your company's budget without checking its work? Probably not. So here's the uncomfortable follow-up question: why would an AI agent recommend your brand to someone else's budget — unless it trusted you enough first?

That question is no longer theoretical. A piece published today in Search Engine Journal by Purna Virji (Principal Consultant at LinkedIn) draws on new Wharton Business School research to reframe the entire AI visibility conversation — and it's a signal every CMO needs to hear right now.

The thesis: trust is the new ranking factor in AI search. Not keywords. Not backlinks. Not even structured data. Trust.

---

For the last 30 years, search optimization has been a conversation between your content and an algorithm. You wrote pages. Google crawled them. Humans clicked.

That loop is breaking.

As AI agents — systems like ChatGPT, Perplexity, Google's AI Overviews, and increasingly autonomous purchasing agents — become the first point of contact between buyers and brands, the question changes from "can this content rank?" to "will an AI agent trust this brand enough to surface it?"

The distinction matters enormously. An AI agent isn't passively surfacing results for a human to evaluate. In agentic commerce scenarios — already live in B2B SaaS procurement, travel booking, and e-commerce — the agent is actively narrowing a shortlist, sometimes making decisions directly.

If your brand doesn't clear the trust threshold, you don't make the shortlist. You simply don't exist in that transaction.

---

The research Virji cites, authored by Stefano Puntoni, Erik Hermann, and David Schweidel from Wharton, identifies three components of trust that determine whether a person relies on an AI agent — and, critically, whether an AI agent will confidently recommend a vendor or brand.

Translated into a marketing lens, these components become an actionable GEO framework:

AI models are trained on the web. That means your competence as a brand is evidenced by the quality, depth, and consistency of what's been written about you — by you and by others.

Thin content, contradictory claims, or a weak third-party mention footprint are trust deficits. A brand that is cited, quoted, and referenced across multiple credible, independent sources sends competence signals that AI models can detect and weight.

GEO implication: Invest in long-form, citable content. Earn mentions in authoritative industry publications. Make your expertise legible to a model that is essentially running a perpetual literature review on every brand it encounters.

AI agents are being designed — explicitly — to reduce uncertainty for the people using them. A brand that consistently puts the customer's outcome first (evidenced through reviews, case studies, transparent pricing, clear refund policies, and community sentiment) reads as benevolent to a model trying to minimize risk for its user.

This is why review management has become a GEO priority, not just a local SEO tactic. And it's why brands that bury pricing, obscure limitations, or generate patterns of negative sentiment are increasingly invisible in AI-mediated recommendations.

GEO implication: Audit your brand's "help vs. sell" ratio across all public-facing content. AI agents weight sources that consistently reduce user uncertainty over sources that create it.

Integrity, in the Wharton framework, is about consistency. AI models have long memories — they're trained on years of data. A brand that has pivoted its messaging repeatedly, made claims that weren't substantiated, or attracted coverage patterns suggesting a credibility gap will carry those signals into AI-generated recommendations.

GEO implication: Consistency across your website, press mentions, social presence, and customer-facing materials is no longer just a brand guideline — it's an algorithmic input.

---

It's tempting to read "agentic commerce" as a future-state scenario. It isn't. Consider:

• Perplexity already surfaces product recommendations with direct purchase links, powered by its own agent layer.
• OpenAI's operator model is designed explicitly for autonomous task completion, including procurement.
• Google's AI Overviews are now appearing in over 47% of commercial queries according to recent tracking data — meaning AI is mediating nearly half of all commercial search interactions in Google's ecosystem.
• Enterprise B2B buyers are already using AI tools to generate vendor shortlists before a human ever enters the evaluation cycle.

The brands that show up on those AI-generated shortlists are the brands that have built trust signals legible to models. The brands that don't — regardless of how strong their traditional SEO is — are experiencing what we're calling AI visibility gaps: the difference between how well a brand performs in traditional search versus how it performs in AI-mediated discovery.

---

The SEJ piece frames the question correctly: the conversation needs to shift from "How do I optimize my website for an LLM?" to "How do I optimize my brand for an autonomous agent?"

Here's where to start:

1. Run an AI brand audit. Query ChatGPT, Perplexity, and Google's AI Overviews with your core customer questions. See whether your brand appears — and how it appears. Are you described accurately? Favorably? At all?

2. Map your trust signal gaps. Using the Wharton framework: where are your competence, benevolence, and integrity signals weak? Is your third-party mention footprint thin? Are review patterns mixed? Is your content contradicting itself across pages?

3. Prioritize citable content creation. Not just blog posts — research, data, original studies, and frameworks that other publications will reference. Citations are the currency of AI trust.

4. Treat review management as a GEO task. The sentiment patterns that AI models extract from your reviews are shaping how they describe you. This is no longer optional maintenance.

5. Audit for consistency. An AI model drawing on 3 years of your brand's public footprint will surface inconsistencies. Audit your messaging, claims, and positioning for coherence over time.

---

The ranking factors of traditional SEO — technical health, backlinks, keyword relevance — don't disappear in the age of AI agents. But they are increasingly necessary-but-not-sufficient conditions.

The brands that win in AI-mediated discovery will be the ones that have made trust legible. Not just to humans, but to the models that are now sitting between your brand and your buyer.

If an AI agent won't confidently recommend you, no amount of keyword optimization will close that gap.

The question isn't whether AI agents are making brand decisions. They already are. The question is whether your brand is built to be trusted by them.

---

Avisible helps brands identify and close AI visibility gaps — the space between how you perform in traditional search and how you're represented in AI-generated answers. If you want to understand your AI trust signal profile, get in touch.