Can a law, consulting or accounting firm license its data for AI training?
Yes, if the firm owns what it licenses and keeps client confidential information out. Every firm holds two kinds of material: what clients entrusted to it, and what the firm built for itself to do the work. The second kind, its procedures, templates, precedents, methodologies and software, is what you can license, and it is easy to overlook as an asset. Much of a company's data already lives in third-party software. Licensing it means re-licensing what already exists, and getting paid for it rather than giving it away for free.
Which professional services data can you license?
The material your firm wrote for itself, rather than for one client, is the core of what you can license:
- SOPs and playbooks: engagement intake, conflict and independence checks, review and sign-off steps, and quality-control checklists (SOPs and process documentation).
- Templates, precedents and know-how: clause banks, model engagement letters, audit work programmes, methodology guides and manuals for in-house tools, ideally with drafting notes (internal documents and knowledge bases; technical documentation and specs).
- Support and service conversations: IT help-desk tickets, internal Q&A threads and process queries, with client facts removed (support tickets and customer conversations; chat and email archives).
- Code: document automation, analytics scripts and client portals, with commit and review history (codebases and engineering history).
Where does this data live in a professional services firm?
Most of it already sits in software your firm pays for: a document management system, a wiki or precedent bank, a practice management system, a help desk, Teams, Slack or email, and a Git host for the code. No new data collection is needed.
Why do AI labs and agent builders want professional services data?
Because law and accounting are among the fields AI developers most want to automate, and your firm's documents show how that work is really done. Reporting on the market for company data, Business Insider wrote in October 2026: "Fields that AI giants are trying the hardest to automate, like law and accounting, see especially high demand" (Business Insider, Oct 13 2026). OpenAI's GDPval benchmark, published in October 2025, builds its tasks from the representative work of industry professionals with an average of 14 years of experience, and its professional, scientific and technical services sector lists lawyers and accountants and auditors among its occupations (GDPval, arXiv 2510.04374).
Data that shows how work is actually done provides "critical context on how decisions are made", an AI training company's chief revenue officer told Business Insider, and that information can help train agents (Business Insider, Oct 13 2026). A firm's own playbooks hold exactly that. The same report described a marketing agency that had licensed files ranging from design files to checklists for onboarding clients, and its owner said it had not sold client-specific material (Business Insider, Oct 13 2026).
What do professional rules say about client information?
They put client information off limits without the client's consent, so the line that matters is between what your firm authored and what your clients entrusted to it. Your own jurisdiction's version of each rule governs.
- Lawyers. ABA Model Rule 1.6(a) says a lawyer "shall not reveal information relating to the representation of a client unless the client gives informed consent", or one of the listed exceptions applies, and Rule 1.6(c) requires reasonable efforts to prevent unauthorized disclosure (ABA Model Rule 1.6). The ABA's comment explains that the rule covers "all information relating to the representation, whatever its source", which is wider than privilege, and that it also reaches disclosures that "could reasonably lead to the discovery" of protected information by a third person (Comment on Rule 1.6). The attorney-client privilege and the work product doctrine are related bodies of law that protect the same relationship (Comment on Rule 1.6), so the safer course is to keep privileged advice and work product out entirely.
- Accountants. The AICPA Code says a member in public practice "shall not disclose any confidential client information without the specific consent of the client" (ET 1.700.001), and its interpretation on third parties says that using such information for the member's own purposes, even without the client being specifically identified, needs the client's specific consent, "preferably in writing" (ET 1.700.060) (AICPA Code of Professional Conduct). The same interpretation does not prohibit advising others based on expertise from prior client work, but if the information may be identifiable to a client, it needs that client's specific consent, preferably in writing. Tax practices have a further gate: US law makes it a misdemeanor for a return preparer to knowingly or recklessly disclose, or use for another purpose, information furnished to prepare a return, subject to listed exceptions (26 U.S.C. §7216).
- Consultants. Client NDAs and engagement letters usually define what is confidential and may assign deliverables to the client, so check them for anything that started on an engagement. Under US trade secret law, disclosing or using a client's trade secret without its express or implied consent is misappropriation if, at the time, you knew or had reason to know you acquired it under a duty to keep it secret or limit its use (18 U.S.C. §1839(5)).
- Firm-authored versus client material. In the US, a work an employee prepares within the scope of employment is a work made for hire, and the employer owns the copyright unless both sides agreed otherwise in a signed written instrument (17 U.S.C. §101; §201(b)). That favours your house templates and precedents, but a precedent cut from a live matter still carries client facts, and a template first built as a client deliverable may belong to that client.
This is general information, not legal advice; see whether it is legal to sell company data to AI.
What is professional services company data worth?
There is no public price list for professional services data. One published reference point: an AI training company announced a programme paying midsize companies $100,000 to $2 million for access to anonymised data (Business Insider, Oct 13 2026). That is a buyer's offer, not a fearofgod.ai price. Partnerships typically range from $50K to $3M+.
What does fearofgod.ai handle when your firm licenses its data?
fearofgod.ai runs the licence through five steps: a partnership form, a conversation, scope and terms, data preparation, then delivery and payment, and you approve the scope before anything is shared. fearofgod.ai de-identifies the data before it goes to buyers. Raw data is deleted according to a schedule agreed with the partner. Your company's name stays confidential until the agreement is fully signed, and a completely anonymous partnership is also possible. Read how to sell data to AI companies for the full process and what goes into a data licensing agreement for AI training before you sign.
Selling professional services company data to AI companies
A law, consulting or accounting firm that owns its professional services company data can sell it to AI companies by licensing it with fearofgod.ai, securely and for long-term revenue. Partnerships typically range from $50K to $3M+. Partners can be paid more than once, through renewals and repeat licences.