AI for law firms & legal advice: § 203 StGB, pleadings, research | meinGPT
How law firms use AI while preserving legal professional privilege (§ 203 StGB): drafting pleadings, legal research with sources, contract and document review, client communication — with real workflows, an example prompt, strict review limits and selection criteria for professionals bound by secrecy.
For Lawyers, firm owners, trainee lawyers and firm management.
- Who it is for
- Lawyers, firm owners, trainee lawyers and firm management
- Impact
- Less time in drafting, research and document review — every legal statement stays reviewed and owned by the lawyer
- Task
- Draft pleadings, run legal research with sources and review documents — while preserving legal professional privilege (§ 203 StGB)
AI in a law firm means using generative AI for the preparatory, recurring tasks of legal work — drafting pleadings, researching legal sources with citations, reviewing contracts and documents, and preparing client correspondence. The decisive frame is legal professional privilege: lawyers in Germany are professionals bound by secrecy under § 203 StGB and may only pass client data to external IT or AI providers if those providers are involved as a 'contributing person' within the meaning of § 203 (3) sentence 2 StGB and the service contract meets the professional requirements of § 43e BRAO (text form, careful selection, a confidentiality obligation — plus, for services that directly serve an individual mandate, the client's consent under § 43e (5) BRAO) — on top of EU processing, a DPA and the assurance that input is not used for training. meinGPT offers law firms exactly this separate confidentiality undertaking under § 203 StGB: in it, SelectCode commits in writing to secrecy as a contributing person under § 203 (3) sentence 2 StGB. The AI supports the preparation; every legal statement is reviewed and owned by the lawyer.
From the task to productive AI use
A firm assistant bundles context (tone of voice, pleading templates, internal precedents as a knowledge base) and task. For pleadings and drafts it turns bullet points and the facts into a structured text proposal, which the lawyer reviews and owns professionally. For legal research it summarises sources — statutes, case law, commentary references — with citations, as preparation, not as binding legal advice from the AI. For document review it structures and summarises uploaded contracts and papers and flags clauses worth checking, without deciding anything legally itself. Client correspondence is produced as a draft for approval. Firm knowledge becomes queryable through a knowledge base. Because several leading models are available behind one interface, you can pick the right one per task; permitted data sources can be connected via connector/API. The prerequisite is an operation that preserves privilege — EU processing, a DPA, no training on input and the provider's written confidentiality undertaking.
- Who it is for
- Lawyers, firm owners, trainee lawyers and firm management
- Impact
- Less time in drafting, research and document review — every legal statement stays reviewed and owned by the lawyer
- Task
- Draft pleadings, run legal research with sources and review documents — while preserving legal professional privilege (§ 203 StGB)
What Law firm / legal advice gets done with AI
Concrete, repeatable flows — from the first prompt to a dependable result.
Draft pleadings and other documents
From the facts and bullet points, the assistant produces a structured text proposal for pleadings, letters or submissions. The lawyer reviews the structure, the argument and every legal statement and owns the final text — the AI supplies the draft, not the legal position. Processing stays in an environment that preserves privilege. Because a pleading directly serves an individual mandate, the client's consent under § 43e (5) BRAO is part of the precondition.
Legal research with sources
As preparation, the assistant summarises the relevant statutes, case law and citations with source references, so the legal review can start faster. This is research support, not binding legal advice from the AI — models can misquote or invent sources, so every citation must be checked against the primary source.
Review contracts and documents
The assistant structures and summarises uploaded contracts and papers, flags clauses worth checking (deadlines, liability, termination) as a prompt, and answers questions about the document's content with a location reference. The legal assessment, negotiation and decision stay strictly with the lawyer — the AI does not replace legal review. The same applies here: where the review directly serves an individual mandate, the client's consent under § 43e (5) BRAO has to be obtained first.
Draft client communication
From bullet points, the assistant drafts factual text for recurring correspondence — status updates, requests for missing documents, scheduling. The lawyer reviews content and tone and approves before it goes out; binding legal statements to clients are made only by the lawyer.
Make firm knowledge accessible
An assistant connected to precedents, internal memos and working instructions answers the team's recurring questions with a source reference — trainees and new colleagues find the right approach faster. Access follows the permissions granted, including with a view to the separation of matters.
Build a firm assistant and bake in the review steps
Firm management assembles an assistant with its own precedents, tone of voice and a fixed review reminder ('double-check every legal statement') without code and releases it to the team — so human control is part of the workflow, not optional.
A real prompt, a real answer
Nothing hidden — you see the input and the result before you sign up.
Summarise this uploaded tenancy agreement in a structured way and flag clauses worth checking: parties, term, notice periods, rent-increase clause, liability and damages provisions. Give locations (section/paragraph in the document) and make no legal assessment — only flag what I as the lawyer should review.
| Point | AI summary | Location | Status |
|---|---|---|---|
| Parties | Landlord … / tenant … | § 1 | captured — please check |
| Term | open-ended, from 01/08/2026 | § 2 | captured — please check |
| Notice period | 3 months to month end | § 12 | ⚠ check: deviation from the statutory period? |
| Rent increase | stepped rent, +2.5 % annually | § 5 | ⚠ check: permissibility / form |
| Liability | liability excluded for slight negligence | § 9 | ⚠ check: standard-terms control |
| Legal assessment | — (deliberately none) | — | ✓ contains no binding statement |
Put it to work in your own company
In a short live demo we show how this solution runs in your company with meinGPT, GDPR-compliant — using your own use cases.
Or download the AI for professionals bound by secrecy — the § 203 selection criteria (PDF, German) as a PDF:
A work email is enough — processed in line with the GDPR.
Built for enterprise compliance
Law firms process professional secrets, and the operation is designed accordingly. meinGPT is operated by SelectCode GmbH, which is ISO 27001 certified and has its security reviewed regularly through independent penetration tests (most recently SySS, 2025). Processing takes place in the EU, a data processing agreement (DPA) is standard, and input is not used to train the models. Access to pleadings, documents and the knowledge base follows strict permissions through central permission management with SSO, is limited by least-privilege scopes and is traceable through audit logs — important also for separating matters and handling conflicts of interest. For professionals bound by secrecy the decisive building block is added: meinGPT offers law firms a separate confidentiality undertaking under § 203 StGB. In it, SelectCode commits contractually and in writing to secrecy as a contributing person under § 203 (3) sentence 2 StGB — going beyond the text form that § 43e (3) BRAO requires for the service contract — keeping entrusted secrets confidential (including beyond the end of the contract), obtaining only the knowledge necessary to perform the contract, and binding all deployed staff and any sub-contractors to confidentiality in writing as well (undertakings to be produced on request). Together with EU processing, a DPA and the exclusion of training on input, that is the basis for using AI while preserving professional privilege; meinGPT provides the confidentiality undertaking on request. Client data therefore stays under control in the firm's environment rather than being processed through private AI accounts.
- § 203 StGB / § 43e BRAO: Does the provider commit contractually to confidentiality as a contributing person — with a separate confidentiality undertaking — and does the service contract meet the text form required by § 43e (3) BRAO?
- § 43e (5) BRAO: Is it settled for which tasks the client's consent is obtained — required as soon as the service directly serves an individual mandate?
- EU processing & DPA: Is client data processed within the EU and is there a data processing agreement (Art. 28 GDPR)?
- No training on input: Is it contractually assured that prompts, pleadings and uploaded documents are not used to train the models?
- Matter separation & permissions: Can access be limited per matter and role (least privilege), so only authorised people see client data — including with a view to conflicts of interest?
- Evidenceability: Is there central permission management, SSO and audit logs to make access traceable?
- Human control: Does the setup support a process in which every legal statement is reviewed and owned by the lawyer?
- Adoption: Is there training and are there champions, so the firm's team uses AI correctly, with review and in line with privilege?
What this solution cannot (yet) do
Honesty is part of the solution. These limits are known — and therefore plannable.
Privilege first: client data may only go into an AI environment in which the provider is bound to confidentiality as a contributing person under § 203 (3) StGB — never through private AI accounts (shadow AI), as that can amount to a criminally sanctioned disclosure.
Every legal statement must be reviewed and owned by the lawyer — the AI may draft, research and review, but must not assess, advise or decide legally on its own.
AI models can misquote or invent sources, case law or citations; every statute and decision cited must be checked against the primary source, and a four-eyes principle is strongly recommended.
The AI does not replace legal advice: binding legal assessment, contract drafting and representation remain the lawyer's responsibility. Conflicts of interest and the separation of matters must additionally be preserved through permissions.
Frequently asked questions
Yes — but only while preserving privilege. Lawyers are professionals bound by secrecy under § 203 StGB. External IT and AI providers count as 'contributing persons'; client data may be made accessible to them under § 203 (3) sentence 2 StGB insofar as this is necessary for the professional activity; professional law then requires, under § 43e BRAO, a service contract in text form with careful selection and a confidentiality obligation — and, for services that directly serve an individual mandate, the client's consent under § 43e (5) BRAO. On top of that you need EU processing, a DPA and the assurance that input is not used for training. meinGPT provides law firms with a separate confidentiality undertaking under § 203 StGB, in which SelectCode commits in writing to secrecy as a contributing person under § 203 (3) sentence 2 StGB. Client data must not be processed through private AI accounts without such a binding commitment.