AI Perception: The New Dimension of Reputation Risk

Large language models like GPT-4o, Claude, and Gemini are increasingly used as information sources by professionals, journalists, and decision-makers. What these models "know" about an entity — and how they present it — has become a new dimension of reputation risk.

The AI Perception Challenge

When someone asks an AI assistant about a person or company, the response is shaped by the model's training data. This creates several risks:

• Outdated information: AI models may reflect data from months or years ago, missing recent developments.
• Inaccurate associations: Models may confuse entities with similar names or incorrectly attribute information.
• Negative bias amplification: Controversial mentions in training data may be disproportionately emphasized.
• Missing context: AI responses often lack the nuance that human judgment provides.

Why AI Perception Matters

Consider these scenarios:

• A potential investor asks ChatGPT about your company before a meeting. The AI provides outdated or incorrect information that colors the conversation.
• A journalist uses AI to research a story about your industry. The AI's characterization of your company shapes the article's narrative.
• An AI-powered screening tool evaluates your executive team. Inaccurate AI representations lead to unfair risk assessments.

How TCI Measures AI Perception

The AI Perception dimension (available in Strategic and Executive tiers) works by:

• Querying AI models about the entity using structured prompts
• Comparing AI responses against verified public data
• Measuring accuracy across key attributes: identity, role, achievements, controversies
• Identifying gaps where AI knowledge is missing or incorrect
• Scoring alignment between AI perception and documented reality

Managing AI Perception Risk

Organizations can take proactive steps to manage how AI models perceive them:

• Maintain a strong, consistent digital presence across authoritative sources
• Ensure Wikipedia and other knowledge bases contain accurate, up-to-date information
• Publish authoritative content that AI models can reference
• Monitor AI perception scores over time to detect shifts

The Trust Capital Index AI Perception module provides the measurement framework needed to understand and manage this emerging risk dimension.