As AI becomes more ingrained in everyday life and spills into client–lawyer relationships, firms are increasingly forced to spend additional time correcting inaccuracies and managing expectations, prompting one firm to raise fees for clients who rely on AI in their correspondence.
As clients grow more comfortable experimenting with AI, the legal profession must continually adapt, balancing efficiency, transparency, and “human” judgment to deliver value in an evolving landscape.
In a post on LinkedIn recently, US-based firm Gerben IP founder Josh Gerben said his firm had an “AI problem”, with clients using AI to write emails to the firm, often with incorrect information about their trademark matter.
This has led to lawyers at the firm spending more time – sometimes twice or three times as much time – on these client matters, identifying and explaining the incorrect information a client has gotten from ChatGPT.
“This has led to a new policy – if we notice a client using AI to communicate with us, we inform the client that the cost of handling the matter will likely be 2-3x [sic] the normal costs (given all the extra time we need to invest),” Gerben said.
Gerben IP offers numerous services on a flat-rate basis, meaning that extra time spent on a matter means the firm has had to raise prices, compared to a firm with strictly billable hours.
Following the post, which, at the time of writing, has more than 2,300 reacts, 511 comments, and 121 reposts, Lawyers Weekly spoke to firms on Australian shores as to whether this is something they’re seeing clients do here – and whether this issue is likely to get worse as AI becomes more commonplace.
Sprintlaw co-founder Alex Solo said the firm is “definitely” seeing clients do this, but that matters aren’t necessarily always impacted negatively.
“It is similar to when clients used to turn up with free template documents they’d found online – sometimes, they were useful in framing the issues, but more often, they created confusion,” he said.
“With AI, it’s similar: sometimes it helps clients synthesise their questions, but other times, it introduces inaccuracies that we need to carefully correct. How much extra time it adds really varies – in some cases, it slows us down, but in others, it can actually speed things up by surfacing the right issues.”
Hicksons partner David Fischl also referenced clients using online templates – but added that he welcomes clients who have “done their own research”, as this often means clients are “more engaged and invested in the process” and adds value to their conversations with the firm.
“Of course, AI outputs can sometimes include inaccuracies or gaps, but that’s not new – clients have always brought in material from Google searches or online templates they’ve found. It’s vital that we, as the lawyers, have an honest conversation with the client about how they achieve the required outcome,” he said.
“If what the client has provided is not useful, then it’s the lawyer’s role to say so clearly. It’s our role as the lawyer to demystify the law and hold our clients’ hands to address their issues and lead them to success.”
When asked if using a legal-specific AI tool could help firms filter out inaccurate information, Fischl said that while these tools are better than stock-standard AI, they’re unlikely to pick up on client mistakes.
“Unfortunately, specialist legal AI tools are currently not a one-stop shop for all legal issues. AI tools that focus on dealing with legal issues certainly have a leg up to general tools like ChatGPT. However, the devil is in the detail for detangling AI-created content, which is often a web of complex, convoluted information that sounds sort of right, but is still wrong,” he added.
“As it stands, unless a subject matter expert for that specific area configured the tool, like we do at Hicksons, it is unlikely a specialist AI tool on the market can adequately solve this problem.”
Solo echoed a similar sentiment, emphasising the importance of legal-specific AI being embedded in workflows.
“Legal-specific AI, when used by lawyers inside a well-designed workflow, can be a huge efficiency boost. But if the client is just using an open-ended chat interface, the same risks apply: the quality depends heavily on the prompt,” he said.
“The real gains come when AI is embedded in client workflows, pre-prompted and quality-controlled, rather than clients experimenting with it in isolation.”
The real risk around this issue, according to CXT Legal director Christos Tsonis, is not AI itself – it’s “clients thinking they can replace judgment with automation”.
“A flawed AI draft (whether presented as instructions or an automated term sheet) in an M&A deal doesn’t save costs; it compounds them,” he said.
“The firms that thrive will be those that draw a clear line between clever prompts and professional accountability and educating clients on when top-tier legal scrutiny is called in and applied, regardless of technology.”
Ongoing adaption alongside AI
For firms that charge clients on a partly or wholly fixed-fee model, spending additional time correcting AI could result in lost revenue – potentially leading to an increase in fees or a disclaimer to clients, similar to Gerben IP’s new policy.
“An important responsibility for lawyers delivering fixed-fee work for clients is to ensure there is alignment on scope. Lawyers need to be honest and direct with their clients. If a client provides commentary or documents that are not helpful to achieve what the client needs, then it is the lawyer’s role to say so before doing any work,” Fischl said.
“For example, if the original scope was to draft an NDA but the client provides an NDA created by ChatGPT, then the lawyer should discuss this with the client and provide them with options. The options could include amending the AI-generated NDA or to start from the firm’s template. Often using the firm’s template is more time- and cost-effective.”
Firms operating on fixed fees can also adapt by incorporating AI as part of their service offering – but Tsonis warned that transparency and alignment are key here.
“Fixed fees are only sustainable if clients understand that AI outputs can’t be treated as plug-and-play legal documents. Every error we untangle is hidden cost. Unless firms reset the conversation and there is alignment, this could quietly become the biggest margin-killer in modern legal practice,” he said.
“AI will only make this issue more common, but forward-thinking firms like ours won’t fear it. The opportunity is to show clients how to use AI intelligently, and to charge for the real value: judgment, risk allocation and the precision that no algorithm can guarantee.”
Similarly, Solo said that as clients engage more with AI personally, legal professionals will need to double down on uniquely human skills while also adapting to the expanding everyday uses of the technology.
“The challenge is to adapt. For fixed-fee firms like ours, this means making AI part of the service: using it to streamline our own processes so we can spend more of our human time on judgment calls and complex strategy. At the same time, part of our role is educating clients about when AI is helpful and when it can mislead. That way, clients feel less need to DIY with generic tools,” he added.
“Just as doctors had to adapt when patients started Googling symptoms, lawyers will adapt to clients using AI. Over time, clients will learn where it adds value and where it falls short. And lawyers who embrace AI, rather than push back against it, will be best placed to deliver value in this new environment.”
Further, lawyers who utilise AI in their own practice will also be more equipped to deal with clients using and experimenting with AI, according to Fischl, who emphasised that lawyers have a “responsibility to find the most time- and cost-efficient way to achieve the client’s outcome”.
“The relationship between lawyers and their clients needs to grow alongside AI. We can pretty safely assume that AI will continue to be used more and more every day. This means more complex issues in a more complex world. Clients will need experienced lawyers to help navigate this,” he added.
“The best lawyers for clients will be those that use AI effectively for their own work, but – never lose sight or focus on the human element of the legal profession – delivering outcomes and strengthening relationships.”
Lauren is the commercial content writer within Momentum Media’s professional services suite, including Lawyers Weekly, Accountants Daily and HR Leader, focusing primarily on commercial and client content, features and ebooks. Prior to joining Lawyers Weekly, she worked as a trade journalist for media and travel industry publications. Born in England, Lauren enjoys trying new bars and restaurants, attending music festivals and travelling.
