How to Get Your Brand Into AI Models

A Conversation with Christopher S. Penn

When users ask ChatGPT or Claude about your company, what do they learn? For many brands, the answer is disappointing—either outdated information or nothing at all. Christopher S. Penn, renowned marketing expert and Chief Data Scientist at Trust Insights, has made understanding AI model training data his specialty.

In this conversation with Maryrose Lyons, Penn reveals how brands can ensure their content appears in AI training datasets—and why podcast guesting is one of the most effective strategies.

The New SEO: Optimising for LLMs

Traditional search engine optimisation focuses on ranking in Google. But as users increasingly turn to large language models for answers, a new discipline is emerging: LLM optimisation. Penn argues that brands must now consider how AI systems represent them.

The challenge is that LLMs are trained on specific datasets. If your content is not in those datasets, the model cannot reference you accurately—or at all. This creates both risk and opportunity for organisations seeking to build authority in the AI age.

How AI Models Learn About Brands

Penn explains that LLMs are trained on vast corpora of text from the web, books, and licensed content. The models develop associations based on patterns in this training data. If your brand appears frequently in high-quality contexts, the model develops accurate, positive associations.

Conversely, if your brand is absent or appears only in low-quality contexts, the model may hallucinate information or omit you entirely when users make relevant queries.

Podcasts: The Secret Weapon

One of Penn's key insights is the value of podcast appearances. Podcast transcripts are increasingly included in AI training datasets. When you appear as a guest on relevant shows, you create authoritative content that models can reference.

Penn practises what he preaches—offering himself as a podcast guest and encouraging others to ask. This strategy builds distributed authority across multiple high-quality sources, increasing the likelihood that AI models will accurately represent your expertise.

Practical Steps for Brand Visibility

Penn recommends several tactics for improving brand presence in AI models:

Create Transcript-Friendly Content: Publish content in formats that are easily processed—clear text, structured articles, and transcripts of audio and video content.

Seek High-Quality Mentions: Pursue guest appearances on established podcasts and publications. Quality of context matters more than quantity of mentions.

Maintain Consistent Messaging: Ensure your key messages are consistent across platforms. Models develop associations from repeated patterns.

Monitor AI Representation: Regularly query major LLMs about your brand to understand how you are currently represented—and address gaps.

Implications for Professional Services

For engineering, architecture, and professional services firms across Ireland and the UK, Penn's insights are particularly relevant. These industries rely heavily on reputation and expertise. As potential clients increasingly use AI to research firms, being accurately represented in model outputs becomes competitive necessity.

Firms should audit their digital presence with AI visibility in mind—not just traditional SEO, but LLM optimisation.

Conclusion

The shift from search engines to AI assistants represents a fundamental change in how brands are discovered. Organisations that understand and adapt to LLM optimisation will have significant advantage. For businesses in Ireland, Dublin, Athlone, and across the UK, the time to start is now—before competitors establish unassailable presence in AI training data.

Want the full conversation? Watch the Chatting GPT episode on YouTube here: https://www.youtube.com/watch?v=W7Bxfba0ol8