Should you use AI to draft your letters of engagement?

In this month’s article from Lockton, Matthew Thomson looks at the key risks of AI-generated letters of engagement.

Generative AI systems have the ability to answer questions and create content at a pace and scale we’ve never experienced before. These new tools have the potential to streamline processes, boost efficiency, and potentially allow lawyers to focus more on strategic tasks. However, as with any new technology, there are important risks to consider, especially since solicitors handle sensitive information and depend on precision and accuracy in their work.

The Law Society of Scotland has a Guide to Generative AI; a detailed document to help the legal profession make informed decisions about how to safely incorporate the use of generative AI products into their legal practice. The guide uses the experiences of law firms and technology experts to provide advice on key issues surrounding the use of generative AI.

Letters of Engagement

In this era of new technology, firms may well be considering where AI might fit into their client-onboarding workflows. Is it a step too far to use AI to draft letters of engagement? On the surface, automating this task could significantly speed up processes and reduce workload. Is this a sensible way to accelerate operations, or could firms be exposing themselves to costly risks that could outweigh the benefits?

The contract between solicitor and client

The letter of engagement is the contract governing the relationship between the solicitor and the client. It is therefore essential that the terms within a letter of engagement are accurate and reflect the scope of work and expectations on both sides. The case of Minkin v Cawdery Fireman & Taylor (2011) EWHC 177 is a reminder of the importance of a robust letter of engagement. In that case, it was held that law firms must be fully aware of the contents of their engagement letters, which need to be in clear terms and complied with by solicitors as well as clients. Although this was an England & Wales case, the requirement for accuracy in letters of engagement is equally relevant here in Scotland. A well-drafted letter of engagement is crucial to manage a client’s expectations from the outset and helps to prevent misunderstandings or disputes down the line.

The Master Policy insurers have identified that the absence of a properly scoped engagement letter is a key factor in successful claims against professionals. A robust engagement letter is therefore not only an essential contractual document but also a risk management tool. It can shield firms from claims and help to mitigate any risk that may arise (see our guide to letters of engagement for some drafting tips).

 The Risks of Using Generative AI Systems for Letters of Engagement

Inaccurate Content

A fundamental risk surrounding the use of AI can be summarised in just one sentence: the information it generates isn’t always correct. Recent cases have brought this issue into sharp focus. One notable example is the case of R (Ayinde) v London Borough of Haringey (2025) EWHC 1040, which serves as a cautionary tale regarding the use of AI. Lawyers in this case were heavily criticised for their “appalling professional misbehaviour” after they relied on citations generated by AI in support of a judicial review application. These citations turned out to be entirely fictitious, undermining the credibility of their submission and highlighting the dangers of unverified AI-generated content.

Earlier this year, there was a similar issue arise in a Scottish rent arrears case. Here, AI invented fake legislation and fictitious references, leading the Sheriff to state that “The lodging of false legal references has the potential to obstruct justice.” These examples highlight the real-world consequences of trusting AI-generated information without thorough verification.

In the context of letters of engagement, any inaccuracy might result in solicitors inadvertently failing to set out to clients what they intend to do for them and how they intend to do it. This would not only have the potential to create confusion but could also mislead the client. Instead of managing and mitigating risk—a key purpose of a letter of engagement — relying on AI-generated drafts without thorough review could introduce a host of unforeseen risks.

With this in mind, solicitors should never trust AI systems to judge their own accuracy. It is important to always check the quality and accuracy of any work produced by an AI system, not least any section of a letter of engagement that has been generated through an AI system. It’s critically important to assess every response provided by a generative AI tool for potential biases or factually inaccurate information.

Confidentiality and Data Protection Issues

Letters of engagement typically contain private details, including the client’s identity, personal data, and the nature of their legal matters. It’s vital that solicitors protect this information.

When working with AI tools, law firms need to be particularly mindful of how generative AI models are trained, and the ways in which any data entered could potentially be disseminated. This means taking a very cautious approach when framing prompts for AI tools to ensure that no sensitive or confidential information is inadvertently exposed.

A key tip for law firms is to thoroughly scrutinise AI providers before adopting or implementing any new system. It’s essential that stringent due diligence is carried out on every AI tool and third-party providers considered for use. The Law Society’s Guide to IT Procurement and the Guide to Generative AI as mentioned above both offer excellent sets of questions that firms can put to providers to help assess their suitability and security measures before proceeding.

It is crucial that the use of any private AI systems aligns with data protection legislation, such as the GDPR, and adheres to the Law Society’s rules. Additionally, as stated in the Law Society’s guidance, lawyers must never share confidential or privileged information with any public generative AI tool, under any circumstances.

Important to Remain Diligent

There are undoubtedly great opportunities for solicitors arising from the use of AI, and many professional services are increasingly turning to generative AI technology to streamline operations and reduce costs. However, as with all new technologies, alongside opportunity comes risk. We should all take care not to overestimate the value of AI tools. There is no substitute for human expertise.

Where a law firm chooses to use an AI tool, there are always key risks to be mindful of. The technology’s outputs may not always meet professional standards for accuracy or confidentiality, and there is the potential for mistakes or oversights that could have serious consequences. Such risks could ultimately lead to Master Policy claims and client complaints. Solicitors should always check their firm's policy regarding the use of AI and make sure that they are fully compliant.

When it comes to letters of engagement, these should be reviewed carefully before being sent to clients (regardless of whether technology solutions have been used to prepare them initially). Similarly, confidentiality and data obligations must always be complied with.

Finally, we would suggest that law firms consider updating their letters of engagement to make it clear when AI technologies are being used. For further details on this aspect, please see the Law Society’s guide.

Ensuring that we remain diligent in this area not only reduces the risk of claims and complaints, but it also safeguards the trust and confidence of clients.

Matthew Thomson is a client executive in the Master Policy team at Lockton.

This is an article provided by Lockton, the official insurance broker responsible for placing and administer the Master Policy of the Law Society of Scotland. Lockton is the direct point of contact for anything related to the Master Policy, including claims, practice changes and general enquiries.