Two recent developments highlight a practical truth for professional-services firms. People are already using AI for legal help, and new legal tech is increasingly being built around approved policy rather than the document alone. For law firms and accounting firms, the opportunity is to design AI workflows that speed up routine work without letting automation set the rules.
Clients are already bringing AI into legal and tax conversations
A new survey reported by Artificial Lawyer found that 26% of 18 to 24 year olds have used AI for a legal problem, and that people often ask chatbots to explain legal jargon, check advice, draft complaints and emails, and provide reassurance. That means firms are already operating in a world where clients may arrive with AI-generated questions, drafts, or expectations.
For law firms and accounting firms, this makes intake and review more important, not less. Custom AI can help teams sort client submissions, identify missing information, and prepare a first-pass summary for the professional, but the firm still needs clear review steps so advice stays accurate and accountable.
The better CLM lesson is policy first, document second
Artificial Lawyer's report on Precisely's Lexnus points to a different model for legal operations. The platform starts with legal policies and playbooks, then uses AI to analyze existing contracts and templates so the rules can be reviewed and approved before they are published.
That approach is useful beyond in-house legal teams. Professional-services firms can apply the same idea to engagement letters, standard client communications, tax questionnaires, and approval routing. The workflow should encode firm policy first, then use automation to move work faster inside those guardrails.
Why custom AI matters more than generic chatbots
The current wave of legal tech also shows that major platforms and large firms are moving deeper into workflow-based tools. Stripe's move into Clerky highlights continued demand for systems that help manage repeatable legal paperwork and startup processes. That is a signal for firms that serve business clients: demand is shifting toward tools that fit real workflows, not one-off prompts.
Generic chatbots can be useful, but they are not enough for regulated or high-stakes work. Firms need custom AI that can use approved templates, route work to the right person, and preserve the firm's standards. The goal is not to replace professionals; it is to reduce friction in routine tasks so staff spend more time on judgment, client communication, and exception handling.
What firm leaders should do next
The tax-preparer sentencing in Illinois is a reminder that misconduct in filing and reporting can carry severe consequences. While AI cannot solve bad behavior, it can support better controls by making review steps more consistent and making unusual inputs easier to flag.
For leaders, the practical move is to identify one narrow workflow where policy is already clear and the process repeats often. From there, build automation that captures the rules, routes exceptions to humans, and leaves a record of what happened. That is the safest way to start with AI in a law firm or accounting firm.
- Start with a workflow that already has clear rules and repeatable steps.
- Use AI to draft, summarize, route, and flag exceptions; keep policy decisions with humans.
- Design around firm-approved playbooks and templates instead of open-ended prompting.
- Treat review and supervision as part of the workflow, not an afterthought.
Sources watched
- 1/4 of Young People Use AI For Their Legal Needs (Artificial Lawyer)
- Precisely Confronts 'Uncomfortable' CLM Truth With Lexnus (Artificial Lawyer)
- Stripe Moves Into Legal Tech With Clerky (Artificial Lawyer)
