Burson Report Reveals 'Credibility Paradox' in Generative AI, Highlighting Gaps in Perception vs. Trust

Burson, a global communications leader that creates value for clients through reputation, has released a new report, "The Credibility Paradox," which reveals discrepancies in how different audiences perceive the credibility of generative AI answers about brands and companies.

This proprietary research and insight evolve the discussion around Generative Engine Optimization (GEO) from a primarily technical effort focused on visibility and sourcing to a strategic reputation-building opportunity centered on credibility.

"In today's zero-click world, LLMs are the new gatekeepers of reputation," said Corey duBrowa, CEO of Burson. "But visibility is not credibility. Being present in these LLMs is necessary, but not sufficient. Our role is no longer just about making our clients visible. It’s about building a robust ecosystem of evidence so that the answers AI constructs are credible to the audiences that matter most. This research is our playbook for turning the credibility paradox into a competitive advantage."

Burson partnered with AI marketing platform leader Profound to collect thousands of reputation-related responses across seven major AI platforms. It evaluated 85 companies based on the eight factors of Burson's "Reputation Capital" framework: Innovation, Creativity, Workplace, Product, Financial Performance, Governance, Social Contribution, and Leadership. Using Burson's proprietary tool "Decipher," co-developed with cognitive AI company Limbik, responses were assigned a "credibility score" for three distinct audiences—general public, opinion leaders, and business decision-makers—generating over 55,000 credibility predictions.

## Key Findings

- **AI Values Proof Over Positioning:** Fact-based claims related to innovation, products, and workplace culture consistently scored higher than those related to more subjective elements like leadership, governance, and social contribution. This suggests the importance of a strong mix of earned, owned, and social content for GEO, as AI heavily weights independent validation from media coverage, reviews, and conversations.

- **Workplace is an Underleveraged Credibility Lever:** Responses related to the workplace environment were rated as most credible, consistent with LLMs' reliance on verifiable sources like reviews on talent platforms and labor-related reporting.

- **Leadership is AI’s Toughest Credibility Test:** Responses to prompts about leadership were consistently rated the least credible across all industries surveyed. Industries that scored higher had evidence rooted not just in executive messaging, but in governance structures, business performance, and external validation.

- **Credibility is in the Eye of the Beholder:** Business decision-makers rated AI-generated responses as 10% more credible on average than the general public, indicating that audience-specific GEO analysis is essential.

These findings form the basis of a framework developed by Burson to help clients build and protect their reputation in all AI-related scenarios. The framework aims to cultivate an ecosystem of independent, credible voices.

"Across the Asia-Pacific region, much of the discussion around AI has centered on whether a brand name appears in a generative AI response, with less focus on whether those answers are accurate, reliable, and credible," said Red G. Surtida, Head of APAC Intelligence & Transformation at Burson. "The real opportunity for organizations lies not just in securing a share of the answer, but in ensuring those answers are evidence-based, backed by credible sources, and compelling to the audiences that matter most."

Steve Rubel, Executive Vice President, Media Insights & Measurement at Burson, added, "GEO began as a visibility challenge quantified by audit reports. This data reveals it has evolved into something more significant: a test of whether the reputation a company has built in the real world is readable, supported, and credible in the AI-mediated environment where audiences are increasingly forming their opinions. Burson’s framework gives communicators a practical path forward, establishing GEO as the new frontier in reputation management."