FCA authorised · FRN 724952 0117 325 0027Quote & buy →
Insurance wiki
Silent AI is AI-related exposure in an insurance policy that neither expressly covers nor expressly excludes artificial intelligence. It matters because cover for an AI claim then depends on general wording, which leaves room for dispute when it counts.
Part of: AI insurance UK
In short
Silent AI means a policy says nothing either way about artificial intelligence. The insurer may be carrying AI risk it never priced, and the policyholder cannot be sure an AI claim is covered. The term comes from silent cyber, which the PRA addressed in supervisory statement SS4/17 in 2017 before Lloyd's required clarity on cyber from 2020. Insurers are now responding to AI by affirming cover, adding exclusions or clarifying it. Businesses should check their wording at each renewal, disclose their AI work and ask for affirmative cover that names it.
Silent AI is the position where an insurance policy says nothing about artificial intelligence: it neither expressly covers AI-related loss nor expressly excludes it. Wordings drafted before generative AI came into everyday use often do not mention it.
Silence cuts both ways:
| Party | What silence means |
|---|---|
| Insurer | It may be carrying AI exposure it never assessed, priced or capped. |
| Policyholder | Cover for an AI claim turns on general words, such as the business description and existing exclusions, and may not be clear until a claim is made. |
Silent AI sits between two clearer positions: affirmative AI cover, where the policy names AI activity or AI-caused loss as covered, and an AI exclusion, where the policy removes it. For the full comparison, see affirmative AI cover vs silent AI.
The term is borrowed from silent cyber. In supervisory statement SS4/17 (July 2017), the Prudential Regulation Authority described non-affirmative cyber risk as "insurance policies that do not explicitly include or exclude coverage for cyber risk", sometimes called silent cyber. It expected insurers to "introduce measures that reduce the unintended exposure to this risk", such as explicit cover, tighter exclusions or specific limits.
In Market Bulletin Y5258 of 4 July 2019, Lloyd's required all policies to be clear on cyber, by exclusion or affirmative cover, starting with first-party property damage policies incepting from 1 January 2020.
Silent AI is the same problem with a newer technology. We are not aware of an equivalent market-wide mandate for AI as at October 2026; the change is showing up wording by wording. See silent cyber and affirmative cover and cyber and silent cyber exclusions in PI policies.
Illustrative scenarios; whether a policy responds depends on its terms and the facts.
| Policy | Illustrative AI event | Why silence matters |
|---|---|---|
| PI for a firm described as "management consultancy" | The firm now builds AI forecasting tools, and one gives a client wrong figures | The insurer may question whether building AI tools was the insured business. See does PI cover AI mistakes? |
| Cyber | Crafted prompts make a customer chatbot reveal personal data | Cover can turn on whether definitions such as "computer system" reach a third-party AI model. See cyber insurance and AI risks |
| Public or products liability | AI-generated advert content leads to a claim, or an AI-controlled machine damages property | Standard US liability forms now offer optional generative AI exclusions; UK wordings vary |
| PI for an accountant | AI-drafted advice relies on a tax rule that does not exist | Usually a professional negligence claim, but a new AI endorsement at renewal could change that. See AI hallucination liability |
Insurers are affirming cover by naming AI activity, excluding AI-related loss, or clarifying cover with conditions or sub-limits.
So a policy that is silent today may not stay silent at renewal. See AI exclusions and clauses in insurance policies.
Silent AI is AI-related exposure in a policy that neither expressly covers nor expressly excludes artificial intelligence. Whether the policy pays for an AI claim then depends on general wording, such as the business description and existing exclusions, which can be disputed.
It is uncertain rather than good or bad. A silent policy may respond to an AI claim, but the insurer can argue the AI work was not insured or not disclosed. Affirmative wording that names your AI work removes much of that argument.
From silent cyber. The PRA's supervisory statement SS4/17 (July 2017) described non-affirmative cyber risk as policies that do not explicitly include or exclude cyber, and Lloyd's later required clarity on cyber from 2020. The market now applies the same idea to AI.
Read the schedule, definitions, exclusions and every endorsement. If none mentions artificial intelligence, machine learning or automated systems, and the business description does not cover your AI work, the policy is probably silent on AI.
Tell us what your AI work is, where you are based, your turnover and the limit you need. It takes a few minutes online, and a broker will come back to you with options. Or call 0117 325 0027.
Start an AI insurance quote Request a callbackApex Insurance Brokers Limited is authorised and regulated by the Financial Conduct Authority. Registered in England and Wales, company number 07014570. This page is general information, not advice on your individual circumstances, and it does not guarantee that cover will be available or on what terms.
