# The Law Firm AI Cost Model: Per-Lawyer Seats vs an Owned Stack > Legal · AI Course · LEG-10 > Source: https://ibl.ai/solutions/legal/course/law-firm-ai-cost-model > Last updated: 2026-08-25 **What legal AI actually costs — per-lawyer licensing against usage-based and self-hosted alternatives, modeled at real firm sizes, with realization impact stated honestly.** ## The Short Answer **Legal AI priced per lawyer assumes every attorney uses it daily, and most partners never log in. ibl.ai has no per-seat pricing and you own all the code and the data, so a 1,000-attorney firm pays for actual usage rather than 1,000 licences, and leaving does not mean abandoning the deployment.** On ibl.ai you own all the code and the data, run it model-agnostic across any LLM, and pay with no per-seat pricing — so you can deploy anywhere, from your own cloud to a fully air-gapped network. [Request Access](https://ibl.ai/contact) · [Explore Legal](https://ibl.ai/solutions/legal) ## Course facts - **Level:** Foundational - **Duration:** 4.5 hours across 8 modules - **Format:** Workshop with spreadsheet modeling - **Modules:** 8 - **Catalog code:** LEG-10 - **Frameworks covered:** TCO analysis, ABA Model Rules ## What is this course about? Legal AI is often priced per lawyer at rates that assume every attorney uses it daily, and most do not. This course models per-lawyer against usage-based and self-hosted alternatives at 50, 200, and 1,000 attorneys, includes the utilization problem, and confronts the realization question the billable hour creates. ## Who is this course for? - Firm CFOs and finance directors - Managing partners and executive committee members - Legal operations leaders - Practice group leaders evaluating tools ### What do I need before starting? - Bring attorney headcount and any current AI vendor quotes - Comfort with a spreadsheet ## What will I be able to do afterwards? - Classify legal AI quotes by pricing shape - Model per-lawyer cost at 50, 200, and 1,000 attorneys - Quantify the utilization problem across partner and associate ranks - Compute a self-hosting break-even for a firm with an IT function - State the realization impact of efficiency honestly ## What does each module cover? ### Module 1 — What pricing shapes does legal AI come in? Per-lawyer, per-matter, usage-based, and owned — and what each assumes about a firm. _(35 min)_ **Objectives** - Classify quotes by pricing shape - Name each shape's hidden assumption - Identify pricing designed to prevent comparison **Topics:** Per-lawyer pricing · Per-matter pricing · Usage-based · Owned deployment **Activity:** Classify three real legal AI quotes and name each hidden assumption. ### Module 2 — What does per-lawyer cost at 1,000 attorneys? Headcount multiplication across firm sizes, including the staff nobody counts. _(45 min)_ **Objectives** - Model per-lawyer cost at three firm sizes - Include paralegals and staff where licensed - Model growth and lateral hiring **Topics:** Headcount multiplication · Staff licensing · Growth modeling · Lateral impact **Activity:** Build the multiplied bill at 50, 200, and 1,000 attorneys. ### Module 3 — How many partners never log in? The utilization problem, which is more pronounced in law firms than in most industries. _(45 min)_ **Objectives** - Measure utilization by rank and practice group - Compute effective cost per active user - Project realistic adoption **Topics:** Utilization by rank · Practice group variation · Effective cost · Adoption projection **Activity:** Measure utilization from a real deployment and compute effective cost per active user. ### Module 4 — How do you model from matter volume? Bottom-up cost estimation from actual work rather than from headcount. _(45 min)_ **Objectives** - Characterize workload by matter type - Estimate consumption from real volume - Model variability across the year **Topics:** Matter type characterization · Volume estimation · Seasonality · Variability **Activity:** Estimate consumption from one practice group's real matter volume. ### Module 5 — When does self-hosting work for a firm? Self-hosted economics for firms with an IT function, and where it clearly does not fit. _(45 min)_ **Objectives** - Model self-hosting cost including operations - Determine the firm size where it becomes viable - Identify where it clearly does not fit **Topics:** Self-hosting cost · Operations requirements · Viability threshold · Poor-fit cases **Activity:** Compute the break-even and state the firm size where it flips. ### Module 6 — What happens to realization? The billable hour question — efficiency that reduces billable time reduces revenue. _(45 min)_ **Objectives** - Model the realization impact of efficiency - Distinguish where efficiency helps and where it costs - Consider the competitive dynamic **Topics:** Realization impact · Billable hour tension · Competitive dynamics · Client expectations **Activity:** Model the realization impact for one practice group under a stated efficiency gain. ### Module 7 — How do you present this to a management committee? A business case for partners who will immediately ask about realization. _(40 min)_ **Objectives** - Structure the case around competitive risk - Address realization head-on - Present sensitivity rather than a single number **Topics:** Competitive framing · Realization discussion · Sensitivity · Objection handling **Activity:** Present your model to a cohort playing a skeptical management committee. ### Module 8 — Building the three-scenario model The workshop module: a complete model at the firm's real headcount. _(45 min)_ **Objectives** - Complete all three scenarios - Run sensitivity on the decisive variables - Produce a one-page committee summary **Topics:** Model completion · Sensitivity · Committee summary · Assumption documentation **Activity:** Complete the model and write the one-page summary. ## What is the capstone project? **Three-scenario legal AI cost model with realization analysis.** Build a complete cost model at your firm's real headcount covering per-lawyer, usage-based, and self-hosted scenarios, with measured utilization, matter-volume-based estimation, a self-hosting break-even, and an honest realization impact analysis. _Deliverable:_ A working model with a one-page management committee summary. ## How are learners assessed? - Utilization measured from a real deployment, not assumed - Realization analysis must state the negative case honestly - Committee presentation assessed on handling the realization objection ## What ships with the course? - **Facilitator guide.** Session-by-session running order, discussion prompts, and the questions that reliably derail a room. - **Learner workbook.** Exercises, checklists, and the templates each module's activity produces. - **Hands-on lab environment.** A sandboxed ibl.ai deployment so exercises run against real agents, not screenshots. - **Assessment bank.** Scenario questions and rubric criteria mapped to each stated learning outcome. - **Source bibliography.** Every primary regulation and standard cited on this page, linked and dated. ## Which AI agents does this course use? - [Billing Time Agent](https://ibl.ai/solutions/legal/agent/billing-time-agent) - [Knowledge Agent](https://ibl.ai/solutions/legal/agent/knowledge-agent) - [Compliance Agent](https://ibl.ai/solutions/legal/agent/compliance-agent) - [Legal Assistant](https://ibl.ai/solutions/legal/agent/legal-assistant) ## Where does the course material come from? Every module is grounded in primary sources — the regulation, standard, or research itself, not a summary of it. Each was resolved at authoring time. - [Model Rules of Professional Conduct](https://www.americanbar.org/groups/professional_responsibility/publications/model_rules_of_professional_conduct/) — American Bar Association. Fee reasonableness constraints framing the realization discussion. - [AI Index Report](https://hai.stanford.edu/ai-index) — Stanford HAI. Inference cost trends underpinning the usage-based modeling. - [Ideas Made to Matter](https://mitsloan.mit.edu/ideas-made-to-matter) — MIT Sloan. Professional services technology adoption economics. - [Transformers documentation](https://huggingface.co/docs/transformers/index) — Hugging Face. Technical reference for the self-hosting cost model. ## Delivery notes Binding guidance for anyone preparing and delivering this course: - Module 6 is the module that determines whether partners engage. Legal AI business cases that ignore realization are dismissed immediately, and the honest analysis is more persuasive than avoiding it. - Module 3's utilization data is hard to get and worth insisting on. Law firm AI utilization is markedly lower than vendors project, especially among senior partners. - Module 5 must state clearly that self-hosting does not fit most firms under a certain size. A model that always favors it is a sales deck. - Use dated current list pricing. Legal AI pricing changes frequently and stale figures discredit the model. - Coordinate with LEG-8 — the realization question appears in both, and the two courses should reach a consistent position. ## Why run AI training on a platform you own? - **You own the course, not a licence to it.** Course content, learner data, and the platform run inside your perimeter — you own all the code and the data. - **Model-agnostic delivery.** Run the course's AI components on any LLM — Claude, GPT, Llama, Gemini, Command — and switch anytime. - **No per-seat training licences.** Usage-based or self-hosted, so cost tracks actual use rather than headcount. - **Deploy anywhere.** Cloud, private VPC, on-premise, or fully air-gapped — including for cohorts that cannot use public AI tools. ## Frequently asked questions ### What does the The Law Firm AI Cost Model: Per-Lawyer Seats vs an Owned Stack course cover? Legal AI is often priced per lawyer at rates that assume every attorney uses it daily, and most do not. This course models per-lawyer against usage-based and self-hosted alternatives at 50, 200, and 1,000 attorneys, includes the utilization problem, and confronts the realization question the billable hour creates. It runs 4.5 hours across 8 modules across 8 modules, at foundational level, and closes with a capstone: Three-scenario legal AI cost model with realization analysis. ### Who should take The Law Firm AI Cost Model: Per-Lawyer Seats vs an Owned Stack? It is written for Firm CFOs and finance directors, Managing partners and executive committee members, Legal operations leaders, Practice group leaders evaluating tools. Prerequisites: Bring attorney headcount and any current AI vendor quotes; Comfort with a spreadsheet. ### Can we run this course on our own infrastructure? Yes. ibl.ai is model-agnostic and deploy-anywhere — cloud, private VPC, on-premise, or fully air-gapped — and you own all the code and the data. Cohort data, submissions, and any material learners upload stay inside your perimeter, which matters for legal teams that cannot send work to a public AI tool. ### How do we get access to The Law Firm AI Cost Model: Per-Lawyer Seats vs an Owned Stack? Request access and we will set it up for your cohort — hosted by ibl.ai, or running against your own deployment. Tell us the group size and timing you need, and whether it should run inside your own perimeter. ### How much does AI training for legal cost on ibl.ai? There is no per-seat pricing — you pay for usage or self-host and pay only for the infrastructure, so a 5,000-person rollout does not cost 5,000 licences. 1.6M+ users across 400+ organizations run the platform this way, including NVIDIA, MIT, and Syracuse University. ## More Legal courses - [AI and Attorney-Client Privilege: The Architecture Question](https://ibl.ai/solutions/legal/course/ai-and-attorney-client-privilege): Whether sending client material to a third-party AI service waives privilege — the confidentiality analysis, the reasonable-efforts standard, and the deployment that avoids the question. - [Verifying AI Legal Research: Never Cite a Hallucination](https://ibl.ai/solutions/legal/course/verifying-ai-legal-research): A verification protocol for AI-assisted research — why fabricated citations happen, how to catch them every time, and the supervision structure that makes it non-optional. - [Contract Review with AI: Redlining, Risk, and Playbooks](https://ibl.ai/solutions/legal/course/contract-review-with-ai): Encode your firm's negotiating positions into an AI review workflow — clause extraction, deviation detection, risk scoring, and where a partner still reads every word. - [AI in eDiscovery: TAR, Privilege Screening, and Defensibility](https://ibl.ai/solutions/legal/course/ai-in-ediscovery): Use AI across the discovery lifecycle while keeping the process defensible — technology-assisted review, privilege screening, validation, and the meet-and-confer record. - [Client Intake and Conflicts Checking with AI](https://ibl.ai/solutions/legal/course/client-intake-conflicts-with-ai): Faster intake without a missed conflict — entity resolution across a matter history, adverse party detection, and why the conflicts decision stays human. - [Ethical AI Use Under the ABA Model Rules](https://ibl.ai/solutions/legal/course/ethical-ai-under-aba-model-rules): A rule-by-rule walk through AI in practice — competence, confidentiality, supervision, fees, and communication — with a firm policy you can adopt.