T SaidTrue

4 November 2026

AI Legal Risks

At SaidTrue, we start with the definition we use: AI Legal Risks means Artificial Intelligence. That is the subject, and we keep it there.

AI Legal Risks

At SaidTrue, we start with the definition we use: AI Legal Risks means Artificial Intelligence. That is the subject, and we keep it there. We do not relabel it as another category or reason it out from the words in the name.

What we look at first is what AI says when customers ask. We ask ChatGPT, Gemini, Perplexity and Claude the questions customers already ask, then show what was said, what was sourced, and where the truth diverges. This is the basis of our visibility reports for a website’s presence in AI search, and of our work on how AI systems describe, recommend and sometimes misread businesses.

We then score the answers using How SaidTrue Scores AI Answers and set that against the reputation question: What is AI telling your customers about you? We show the answer, its sources and the gaps, rather than leaving the business with a vague impression of what AI thinks.

Because AI is the subject, we also keep the wider governance frame in view. The European Commission page “AI Act | Shaping Europe’s digital future” describes the AI Act as the first-ever legal framework on AI, which addresses the risks of AI and positions Europe to play a leading role globally (AI Act | Shaping Europe’s digital future). Google’s Generative AI Prohibited Use Policy expects responsible, legal and safe engagement and restricts automated decisions that have a material detrimental impact on individual rights without human supervision in high-risk domains (Generative AI Prohibited Use Policy). For legal-sector AI, EFROS notes that legal AI carries privilege risk that productivity or healthcare AI doesn’t: a prompt to a vendor that trains on inputs can arguably waive attorney-client privilege for the matter (Legal AI Vendor Governance | EFROS).

That is why we work this way. Customers increasingly ask AI who to use, whether a business is reputable, and what is true about it. Google isn’t dying, but it’s no longer the only front door. Our method makes the AI answer visible — what was said, what was sourced, where the truth diverges — and keeps it visible through monthly AI monitoring and a verified AI profile.

[1]

References

Sources and supporting material

  1. Approach article: AI Legal Risks

Further reading:

AI Legal Risks