FCA authorised · FRN 724952 0117 325 0027Quote & buy →
Aipedia: Marketing and PR agencies
AI can draft a press release in seconds, but the agency and client remain responsible for every fact, quote and claim it contains. In PR, one inaccurate statement about a third party can lead to a defamation claim.
Part of: AI in Marketing and PR agencies
In short
PR teams may draft releases, quotes and crisis statements with AI, but models invent details and misattribute facts. The CIPR's June 2026 guide on responsible AI use sets professional expectations, product releases can fall under ASA rules and the DMCC Act ban on misleading actions, and statements about real people engage defamation law. Announcements for listed clients add market disclosure sensitivity. PR consultancy PI may include or offer defamation cover, subject to the wording, but deliberate falsehoods and regulatory fines usually fall outside it.
PR consultants use a generative AI chatbot to turn briefing notes into a draft release, produce holding statements, write spokesperson quotes, summarise background on journalists and adapt one announcement into several formats. In a crisis, AI can produce a first draft statement in seconds. The risk is that a model fills gaps with plausible but false detail: an invented figure, a wrong job title, a quote attributed to the wrong person, or a confident statement about a competitor, regulator or individual that is simply untrue.
The CIPR published The Responsible Use of AI in PR in June 2026, a members-only best practice guide translating the Global Alliance's Venice Pledge, which the CIPR has signed. It sits alongside the CIPR's AI and digital ethics playbook. These set professional expectations for accuracy, transparency and human judgement rather than legal rules.
Where a release promotes products, the ASA's media-neutral approach and section 226 of the DMCC Act on misleading actions can apply. The IPA/ISBA principles also expect fact and permission checks (principle 10) and client agreement before AI content is used (principle 11). Statements about identifiable people or companies engage the general law of defamation, and announcements by listed clients may be subject to market disclosure rules, which makes accuracy and timing especially important.
Illustrative scenario: a PR agency uses AI to draft a client's statement responding to a supplier dispute. The draft says the supplier "was found to have falsified safety records", a phrase the model inferred from a news article about a different company. Under deadline pressure, the account lead sends it out. The supplier issues a defamation claim against the client and the agency, and the client's share price moves on the news, prompting questions about the accuracy of what was released.
Speed is the common factor. Crisis work compresses review time, and an AI draft that reads fluently is easy to approve without checking where each fact came from. Recording sources for every factual sentence, even in a hurry, is a habit that can prevent many of these incidents.
Less dramatic errors can include wrong financial figures, embargo dates set incorrectly, invented customer names in case studies, and quotes approved by nobody at the client.
PI for PR consultancies may include defamation cover, or it may be available as an extension; the scope varies, and some policies exclude claims where the insured knew a statement was untrue. A client's own claim that the agency negligently released inaccurate information may fall under the PI insuring clause, subject to the wording. Regulatory consequences for a listed client are a separate matter, and fines are generally uninsurable. See defamation and brand reputation risk for PR consultants, does PI cover defamation? and PI or media liability for AI claims.
When releases are monitored after publication, see AI media monitoring and sentiment analysis; when they are pushed to social channels, see AI social media management. The marketing and PR agencies hub covers the wider rules.
How Apex helps: Apex arranges PI for PR consultancies. A broker reviews your defamation and media cover and how the policy treats AI-drafted releases and statements.
Potentially everyone involved in publishing it, which can include the client and the agency. Using AI is not a defence. Whether the agency's PI responds depends on the defamation cover in the policy and how the claim is framed.
The CIPR's June 2026 guide, The Responsible Use of AI in PR, is members-only and translates the Venice Pledge into practice. Members should consult the guide for its detailed expectations on disclosure.
It is risky. Confidential or price-sensitive information entered into a public tool may be retained or used outside your control. IPA/ISBA principle 6 and UK GDPR also apply where personal data is involved. Use approved tools with suitable terms, or keep it out.
Not always. Some PI policies for PR consultancies include defamation, but some offer it as an extension or apply sub-limits and exclusions for statements known to be untrue. Check the wording before relying on it for AI-drafted statements about third parties.
Tell us about your firm and the work you do. A broker will review how your current PI treats AI-assisted work, or arrange new cover. No fee for a quote. Or call 0117 325 0027.
Start a PI 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 about how AI is used in the profession and how insurance may respond. It is not legal, regulatory or insurance advice on your circumstances, and it does not guarantee that cover will be available or on what terms.
