In AI-powered search, being visible is no longer enough to earn trust. Burson’s latest view of Generative Engine Optimization, or GEO, argues that brands now face a widening gap between appearing in AI answers and being seen as credible.
That gap has a name: The Credibility Paradox. As zero-click search becomes more common, AI systems and Large Language Models are increasingly acting as the new gatekeepers of brand reputation, which means visibility alone does not guarantee confidence.
Credibility now depends on proof, not just presence
Corey duBrowa, CEO of Burson, said the job of public relations is shifting. Instead of focusing only on making clients visible, communicators now need to build an ecosystem of evidence that can stand up to AI scrutiny.
Burson’s research suggests that AI does not weigh every reputation signal equally. Facts that are easier to verify, such as products and innovation, tend to carry more weight than more subjective claims, including leadership strength.
That finding matters because it changes how brands should think about GEO. Strong messaging on its own is not enough if it is not supported by independent validation.
External signals matter more than brand self-description
The study points to a clear need for more outside proof. Burson says brands should increase the amount of independent validation available to AI systems, including external reviews, organic media coverage, and public discussion.
Those signals help create a stronger factual base for AI-generated answers. Without them, even a highly visible brand may struggle to be perceived as credible.
How the research was built
To produce the report, Burson worked with Profound, an AI marketing platform, and tested thousands of reputation-related queries across seven major AI answer platforms.
The analysis used an entity-based approach to evaluate reputation more comprehensively across platforms. Burson also assessed 85 companies across eight reputation capital pillars: Innovation, Creativity, Workplace, Product, Financial Performance, Governance, Social Responsibility, and Leadership.
In addition, Burson used Decipher, a tool developed with Limbik, to project more than 55,000 credibility forecasts. The responses were also grouped into three audience segments: the General Public, Elite Opinion, and Business Decision-Makers.
Workplace signals stand out
One of the strongest findings involved workplace reputation. For the general public, workplace was the most trusted aspect, even though it is often overlooked in many PR efforts.
Burson says companies should make better use of digital assets tied to recruiting, employment reviews, and independent worker audit reports. These materials can strengthen the evidence base that AI systems read when forming answers.
Leadership claims are weaker than expected
The report also shows that leadership claims perform poorly in AI evaluations. Across most industries, statements about leadership received the lowest trust scores unless they were backed by strong governance structures.
Technology and Aerospace were the exceptions, where governance appeared to support credibility more effectively. The result suggests that leadership-centric campaigns may lose trust if they rely too heavily on executive messaging without hard business proof.
Business audiences are more receptive
The Credibility Paradox also found that perception varies by audience. Business Decision-Makers rated AI answers as 10% more convincing on average than the general public.
That difference means GEO strategies cannot be one-size-fits-all. Brands need to think separately about investors, consumers, and regulators if they want AI-mediated answers to work in their favor.
What this means for PR teams
In Asia Pacific, Burson sees the same pattern emerging. The debate around AI search is often focused on whether a brand appears in the answer, rather than whether the answer is accurate.
Steve Rubel, EVP Media Insights & Measurement at Burson, said GEO now tests whether a company’s real-world reputation matches the digital validation mediated by AI. That makes integrated, cross-channel content more important, since reputation signals need to work together rather than sit in separate programs.
Red Surtida, Head of Intelligence & Transformation APAC at Burson, said the opportunity is not only to secure share of answer, but to ensure those answers are grounded in evidence. That shift places more pressure on brands to build credibility that can be independently verified, not just repeated.
