Legal departments across Southeast Asia are under constant pressure to do more with less. In-house counsel at mid-size companies in Malaysia, Singapore, and Indonesia routinely manage 200+ contracts per year, handle regulatory queries from multiple jurisdictions, and draft everything from NDAs to board resolutions — often with lean teams.
AI tools like ChatGPT, Claude, and Microsoft Copilot can transform how legal teams work. But the legal function carries unique risks: client privilege, professional responsibility, regulatory compliance, and the absolute requirement for accuracy. A generic AI course does not address these concerns. Legal teams need training that is built around their workflows, their governance requirements, and their professional obligations.
Pertama Partners' ARBITER programme (AI for Legal Teams) is a 2-5 day programme designed specifically for in-house legal teams and law firms across the region. It covers the practical AI skills that save time on contracts, research, and compliance — while embedding the governance guardrails that protect the organisation and its clients.
Understanding AI in the context of legal work — what it can do well, where it fails, and why legal teams must approach it differently from other departments.
Contract review is the single largest time sink for most in-house legal teams. AI can reduce review time by 60-75% while improving consistency.
Practical exercise: Participants review a sample vendor agreement using AI-assisted prompts, comparing results with manual review to calibrate accuracy expectations.
AI dramatically accelerates the research phase of legal work — from regulatory questions to precedent analysis.
AI excels at creating first drafts of legal documents that legal professionals then refine. Time savings of 65% on initial drafts are typical.
Key principle: AI generates the structure and initial language; the legal professional applies judgment, verifies accuracy, and ensures compliance with applicable law.
Compliance teams within legal departments manage an expanding regulatory landscape. AI accelerates documentation without compromising thoroughness.
Staying ahead of regulatory changes is essential for in-house teams. AI helps systematise this process.
Legal teams face the strictest governance requirements of any department. This module covers the professional and ethical obligations specific to legal AI use.
| Governance Area | Rule | Rationale |
|---|---|---|
| Client privilege | Never input privileged communications or client-identifiable information into AI tools | Legal professional privilege / attorney-client privilege protection |
| Confidentiality | Never input client names, case details, or sensitive matter information | Professional duty of confidentiality |
| Professional responsibility | All AI outputs must be reviewed and verified by a qualified legal professional | Professional conduct rules across all jurisdictions |
| Accuracy obligation | Every legal citation, statute reference, and case reference must be independently verified | AI hallucination risk — fabricated citations are a documented problem |
| Data protection | Comply with PDPA (MY/SG) and UU PDP (ID) when processing any personal data with AI | Regulatory compliance obligation |
| Disclosure | Follow firm/company policy on disclosing AI use to clients and counterparties | Transparency and professional ethics |
| Record keeping | Document when and how AI was used in legal work product | Audit trail and professional accountability |
Case study discussion: Real-world examples of AI hallucinations in legal filings (fabricated case citations) and lessons for Southeast Asian legal teams.
Professional conduct considerations by jurisdiction:
Hands-on session where participants build a reusable prompt library for their specific practice area.
| Task | Without AI | With AI | Time Saved |
|---|---|---|---|
| NDA first draft | 2-3 hours | 30-45 min | 70% |
| Contract review (key terms extraction) | 3-4 hours | 45-60 min | 75% |
| Legal research memorandum | 4-6 hours | 1.5-2 hours | 65% |
| Compliance checklist creation | 2-3 hours | 30-45 min | 75% |
| Board resolution drafting | 1-2 hours | 20-30 min | 70% |
| Regulatory update summary | 2-3 hours | 30-45 min | 75% |
| Terms of Service first draft | 4-6 hours | 1-2 hours | 70% |
| Employment contract template | 3-4 hours | 45-60 min | 75% |
| Tool | Legal Use Case | Why It Matters |
|---|---|---|
| ChatGPT | General drafting, research queries, compliance documentation | Most versatile; strong at structured legal documents |
| Claude | Long document analysis, nuanced legal writing, contract review | Excels at handling lengthy contracts and producing careful, qualified legal language |
| Microsoft Copilot | Email drafting, Teams meeting summaries, Word document editing | Integrates directly into the legal team's existing M365 workflow |
| Format | Duration | Best For | Group Size |
|---|---|---|---|
| Full Legal AI Programme | 2 days (16 hours) | Complete in-house legal team upskilling | 10-20 |
| Intensive Workshop | 1 day (8 hours) | Core skills — contracts, research, drafting | 10-25 |
| Contract Review Focus | Half day (4 hours) | Teams with high contract volumes | 10-20 |
| General Counsel Briefing | 2 hours | GC, Head of Legal, Legal Directors | 5-15 |
| Legal AI Champions | 3 days | Legal team members who will train others | 5-10 |
| Data Category | Can Use with AI | Conditions |
|---|---|---|
| Template clauses and standard language | Yes | No client-specific information |
| Anonymised contract structures | Yes | Remove all party names, deal values, identifying details |
| Public regulatory text | Yes | Verify currency and accuracy of AI summary |
| Client-specific matter details | No | Privilege and confidentiality obligation |
| Privileged communications | No | Absolute prohibition |
| Personal data (names, IC numbers, addresses) | No | PDPA / UU PDP compliance |
| Draft legal opinions | Conditional | Only with enterprise AI tools with appropriate data handling agreements |
| Metric | Before Training | After Training |
|---|---|---|
| Contract review turnaround | 2-3 business days | Same day |
| First draft quality (usable without major rework) | 1 in 4 AI attempts | 3 in 4 AI attempts |
| Time spent on routine legal research | 4-6 hours per query | 1-2 hours per query |
| Compliance documentation backlog | Growing | Under control |
| Legal team capacity (matters handled per month) | Baseline | 25-40% increase |
| Time to produce regulatory summaries | 2-3 hours | 30-45 minutes |
Legal AI training delivers measurable return on investment. Consider a legal team of 5 professionals, each saving 8 hours per week on drafting, research, and documentation:
Beyond direct time savings, legal teams report improved consistency in document quality, faster turnaround on internal client requests, and reduced reliance on external counsel for routine documentation — each of which generates additional cost savings.
Is it safe for legal teams to use AI tools like ChatGPT? Yes, with strict governance. The critical rules are: never input privileged communications, never include client-identifiable information, never rely on AI for legal citations without independent verification, and always have a qualified legal professional review every output. The course teaches these boundaries in detail and provides a governance framework tailored to legal teams.
Will AI replace lawyers? No. AI accelerates the drafting, research, and documentation phases of legal work — tasks that consume 60-70% of a legal professional's time. It does not replace legal judgment, client relationships, negotiation skills, or professional accountability. The most effective legal professionals will be those who use AI to handle routine work faster, freeing time for higher-value advisory and strategic work.
How does AI handle multi-jurisdictional legal work across Southeast Asia? AI tools are useful for comparing regulatory frameworks across Malaysia, Singapore, and Indonesia — for example, summarising the differences between data protection regimes or employment law requirements. However, jurisdiction-specific legal advice must always be verified by a qualified professional in that jurisdiction. The course teaches how to structure multi-jurisdictional research prompts effectively.
Can law firms use this training for client-facing work? Yes. Law firms are increasingly using AI for first-draft production, research acceleration, and document review. The governance module specifically addresses client disclosure obligations and professional responsibility rules. Pertama Partners' ARBITER programme can be customised for law firm workflows, including matter management and knowledge management integration.
Yes, with strict governance. Never input client-privileged information, confidential case details, or personally identifiable data into public AI tools. Use AI for template creation, research frameworks, and draft structures. Always apply professional legal review to all AI outputs.
A legal-focused AI course covers: AI-assisted contract review and analysis, legal research techniques with AI, compliance documentation automation, regulatory monitoring, and strict governance for handling legal data with AI tools.
