Professional ethics
Legal professionals remain responsible for their work and must exercise knowledge, skill and professional judgment.
Updated 23 July 2026 · AI and legal practice
Artificial Intelligence and the Law in Singapore: Can AI Replace a Lawyer?
AI can assist with research, organisation and first drafts. A court user remains responsible for checking the law, evidence, citations and every document submitted.
Book a Free Initial Consultation
Can AI replace a lawyer?
AI can help with research, organisation and early drafting. A lawyer remains responsible for applying current law to the evidence, identifying procedural and strategic risks, advising on consequences and presenting the client’s case.
The client question
AI can help organise information and generate a starting draft, but its output may be incomplete, outdated or invented. In Singapore court proceedings, the person using it remains responsible for what is filed or submitted.
For clients, the attraction is easy to understand. Generative AI can produce information quickly, organise large amounts of material and reduce the time spent on repetitive legal work. This raises an important question:
Can a person use artificial intelligence instead of engaging a lawyer?
The answer is more nuanced than either complete acceptance or outright rejection.
AI can be a valuable legal tool. It may help a person understand basic legal principles, organise facts and identify possible issues. However, AI-generated information must still be checked against authoritative legal sources, the evidence and the particular circumstances of the case.
Justin Chan, Managing Partner of Justin Chan Chambers LLP, discussed these issues in an interview on Money FM 89.3’s Money and Me, including how AI may affect litigants, lawyers, legal fees, access to justice and the future of advocacy.
How is artificial intelligence being used in Singapore’s legal system?
The Singapore Courts have been exploring the use of generative AI to improve access to justice, particularly for people who represent themselves.
The Singapore Judiciary has developed a generative AI-powered case summarisation tool in collaboration with Harvey AI for users of the Small Claims Tribunals. The tool is intended to summarise case documents for Tribunal Magistrates and self-represented parties, helping users navigate the claims process more effectively.
The Courts have also issued a guide governing the use of generative AI by court users. The guide applies to matters in the Supreme Court, State Courts, Small Claims Tribunals, Employment Claims Tribunals, Community Disputes Resolution Tribunals and Family Justice Courts.
These developments reflect a broader movement towards using technology to make legal processes faster, more understandable and more accessible.
What can AI help a litigant do?
A self-represented litigant may use AI to assist with preliminary tasks such as:
- identifying the general area of law involved;
- understanding basic legal terminology;
- organising a chronology;
- summarising correspondence;
- identifying disputed facts;
- preparing a list of documents;
- structuring a claim or defence;
- preparing questions for a lawyer; and
- obtaining a preliminary overview of possible legal issues.
In his Money FM interview, Justin explained that AI is particularly useful for establishing the groundwork or “first principles” of a case.
For example, where a dispute concerns the sale of an item for which payment was not made, AI may help identify basic contractual concepts such as offer, acceptance, consideration, breach and remedies.
That preliminary information can make a consultation more efficient because the client arrives with a clearer understanding of the possible legal issues.
However, the information should then be checked by a qualified lawyer.
Can AI provide legal advice?
AI can provide general legal information, but it should not automatically be treated as reliable legal advice.
Legal advice requires more than stating general principles. A lawyer must consider:
- the client’s exact facts;
- the available evidence;
- the applicable legislation;
- current case law;
- procedural rules;
- limitation periods;
- evidential weaknesses;
- possible counterclaims;
- commercial or family consequences;
- settlement options; and
- the practical likelihood of success.
Two cases may appear similar at first glance but produce very different results because of a single document, communication, deadline or factual distinction.
AI may identify the general legal framework. It may not reliably determine how the law applies to a particular person’s evidence and objectives.
What are the risks of relying on AI for legal work?
AI may invent cases or legal citations
Generative AI systems can produce answers that sound persuasive but contain non-existent judgments, incorrect quotations or inaccurate statutory references.
This is sometimes called an AI hallucination.
A legal document may appear polished while relying on authorities that do not exist or do not support the proposition stated.
Court users remain responsible for the accuracy of material they submit, even where it was generated using AI.
AI may rely on outdated law
Legislation, court procedures and legal principles change.
An AI response may rely on an earlier version of a statute, an outdated procedural rule or a judgment that has subsequently been distinguished or overturned.
AI may misunderstand the evidence
AI does not automatically know whether a witness is credible, whether a document is authentic or whether an account is contradicted by other evidence.
It may accept the user’s version of events without testing weaknesses in that account.
AI may overlook procedural requirements
A claim can fail even where the underlying complaint appears justified.
Procedural issues may include:
- filing in the wrong court;
- suing the wrong party;
- missing a limitation period;
- failing to plead an essential fact;
- failing to comply with a court direction;
- relying on inadmissible evidence; or
- seeking a remedy that the court has no power to grant.
Confidential information may be exposed
Users should be cautious before placing privileged, commercially sensitive or personal information into publicly available AI systems.
Legal disputes frequently involve private financial records, medical information, family communications, company documents and confidential correspondence.
Can AI reduce legal fees?
Potentially, yes.
Legal fees are often affected by the time required to:
- review documents;
- prepare chronologies;
- conduct initial research;
- summarise evidence;
- compare versions of documents;
- prepare first drafts; and
- organise large quantities of information.
AI may reduce the time required for some of these tasks.
In his interview, Justin explained that work which might once have required several hours could sometimes be completed more quickly where AI performs the initial groundwork.
This does not mean that the output should be accepted without review.
A lawyer must still:
- verify the legal authorities;
- compare the output with the evidence;
- test the reasoning;
- identify omissions;
- correct errors;
- apply professional judgment; and
- assume responsibility for the final work.
The real value of legal AI lies in improving efficiency, not eliminating quality control.
Will AI make litigation more affordable?
AI may help reduce the cost of some litigation tasks, particularly where a case involves repetitive work or substantial documentary material.
It may also help clients prepare for consultations by organising:
- key dates;
- relevant correspondence;
- disputed transactions;
- financial records;
- questions requiring advice; and
- the outcome they wish to achieve.
A well-organised client can reduce the time a lawyer spends identifying basic facts.
However, litigation costs are not determined solely by drafting and research.
Costs may also arise from:
- factual disputes;
- expert witnesses;
- discovery and document production;
- affidavits;
- interlocutory applications;
- hearings;
- cross-examination;
- appeals; and
- the conduct of the opposing party.
AI cannot remove the cost created by a genuinely complex or heavily contested dispute.
Can AI help people in the Small Claims Tribunals?
The Small Claims Tribunals are a suitable environment for testing AI-assisted legal processes because parties generally represent themselves.
AI may help users:
- explain their claim more clearly;
- identify the documents supporting their position;
- distinguish relevant from irrelevant facts;
- summarise lengthy records;
- understand what the opposing party is disputing; and
- present information in a more structured form.
The Judiciary has described the use of AI in the Small Claims Tribunals as part of an effort to assist self-represented persons who may otherwise struggle with unstructured information and unfamiliar court processes.
AI does not decide whether a witness is telling the truth or whether a document should be accepted. Those matters remain for the Tribunal Magistrate.
Will easier access to AI lead to more frivolous claims?
One concern is that people may use AI to generate claims quickly and flood the courts with weak or unmeritorious cases.
In his interview, Justin observed that frivolous claims can already be commenced without AI. The difference is that AI may reduce the effort and expense involved in preparing them.
However, Singapore’s court system contains safeguards.
A party who brings an unmeritorious claim may face:
- dismissal of the action;
- striking-out orders;
- adverse costs orders;
- wasted expenditure;
- procedural sanctions; and
- possible consequences for abuse of process.
The fact that AI made a claim inexpensive to prepare does not make the claim legally sound.
A claimant may save money preparing a document but later be ordered to pay the opposing party’s legal costs.
Can AI replace courtroom advocacy?
AI may help prepare a case, but courtroom advocacy involves distinctly human skills.
An advocate must:
- understand the client’s priorities;
- decide which arguments matter most;
- respond to questions from the judge;
- assess the atmosphere in court;
- cross-examine witnesses;
- adapt when evidence changes;
- make strategic concessions;
- identify the strongest route to a practical outcome; and
- present the client’s case persuasively and responsibly.
In the interview, Justin identified empathy and the commitment to present a client’s case at its highest as qualities that AI cannot reproduce.
Litigation is not simply an exercise in finding legal propositions.
The court must be persuaded that those propositions apply to the evidence and justify the order sought.
Where both sides have access to similar AI tools, the quality of the advocate’s judgment, preparation and presentation may become even more important.
Why does empathy matter in legal representation?
Clients often approach lawyers during difficult periods in their lives.
A legal dispute may involve:
- the future of a business;
- access to children;
- the division of family assets;
- the loss of a home;
- allegations of misconduct;
- the breakdown of a partnership;
- an inheritance dispute; or
- substantial financial exposure.
A lawyer must understand not only what relief is legally available but also what outcome is meaningful to the client.
An AI system may produce a list of possible remedies. It does not experience the financial, emotional or reputational consequences of pursuing them.
Empathy helps a lawyer determine:
- which battles are worth fighting;
- which issues should be prioritised;
- whether settlement is preferable;
- how evidence should be presented;
- what risks the client can tolerate; and
- how the case affects the client beyond the courtroom.
Will AI replace junior lawyers?
AI is likely to change the work performed by junior lawyers.
Tasks such as basic research, document summarisation, initial drafting and large-scale document review may increasingly be supported by AI.
This may mean that lawyers spend less time on repetitive groundwork and more time on:
- legal judgment;
- client advice;
- strategy;
- negotiation;
- witness preparation;
- advocacy;
- cross-examination; and
- verification of AI-assisted work.
Justin noted during the interview that law firms may eventually require less manpower for certain back-end litigation tasks, while lawyers remain essential at the strategic and advocacy stages.
The likely outcome is therefore not the disappearance of lawyers but a change in what clients expect lawyers to do.
How should lawyers use AI responsibly?
Responsible use of AI requires a lawyer to remain accountable for the final product.
Good practices include:
- checking every cited case and statutory provision;
- reviewing primary legal sources;
- verifying quotations;
- protecting confidential information;
- understanding how the tool handles data;
- testing the output against contrary arguments;
- identifying assumptions;
- keeping human control over strategic decisions; and
- avoiding misleading claims about the capabilities of the technology.
Singapore’s Ministry of Law published a guide in March 2026 addressing responsible, ethical and effective use of generative AI in the legal sector. The guide covers lawyers, in-house counsel, allied legal professionals, legal technology providers and others handling legal work.
Confidentiality
Client data and confidential information require protection, including attention to how an AI model processes and stores data.
Transparency
Disclosure may be appropriate where AI use is substantial, affects cost or involves data handling that may concern the client.
How does Justin Chan Chambers use AI?
Justin Chan Chambers LLP uses AI and legal technology to support—not replace—the work of its lawyers.
Depending on the case, technology may assist with:
- organising evidence;
- developing chronologies;
- reviewing documents;
- identifying inconsistencies;
- conducting preliminary legal research;
- summarising complex records;
- preparing internal working drafts;
- managing deadlines; and
- improving client communication.
Every material legal conclusion must still be checked against the evidence, current law and the client’s objectives.
The firm’s approach is to use technology to reduce avoidable administrative work so that lawyers can devote greater attention to strategy, advocacy and client engagement.
Should I use AI before speaking to a lawyer?
AI may help you prepare for an initial consultation.
You may use it to produce:
- a factual chronology;
- a list of the parties;
- a summary of the dispute;
- a list of relevant documents;
- questions you wish to ask; and
- the practical outcome you hope to achieve.
You should avoid assuming that an AI-generated legal conclusion is correct.
If you discuss the output with a lawyer, useful questions include:
- Is this legal principle accurate?
- Does it apply to my facts?
- What evidence is missing?
- What arguments could the other side raise?
- Is the proposed remedy available?
- What are the risks and likely costs?
- Is there a better commercial or family solution?
AI can make the consultation more focused. It should not prevent you from obtaining advice where the consequences are significant.
Watch Justin Chan’s Interview on AI and the Future of Law
Justin Chan appeared on Money FM 89.3’s Money and Me to discuss how generative AI is reshaping legal practice in Singapore.
The interview covers:
- the use of AI by self-represented litigants;
- AI in the Small Claims Tribunals;
- whether AI can reduce legal costs;
- the risk of fabricated legal authorities;
- frivolous AI-generated claims;
- the continuing importance of lawyers;
- empathy and courtroom advocacy;
- the future role of junior lawyers; and
- how AI may affect litigants, law firms and the Judiciary.
Watch or listen to the interview here: https://youtu.be/0AGN2o8uxng?si=KpSpQRVgoebSilGG
The discussion provides a practical introduction for clients who want to understand how AI may reduce legal costs without replacing professional judgment and human advocacy.
Frequently asked questions
Can ChatGPT write a court document?
It can produce a draft, but the document may contain incorrect law, invented authorities, missing facts or procedural defects.
Any court document should be carefully reviewed before filing.
Can I rely on AI-generated legal cases?
No case citation should be relied upon until it has been checked against an authoritative legal database or the official judgment.
Is it permitted to use AI in Singapore court proceedings?
Court users may use generative AI, but they remain responsible for the accuracy and propriety of the material submitted. The Singapore Courts’ guide on generative AI applies across the Supreme Court, State Courts and Family Justice Courts.
Will AI make lawyers cheaper?
AI may reduce the time required for certain tasks. The effect on fees depends on the complexity of the dispute, the volume of evidence and whether hearings, experts or contested applications are required.
Can AI assess whether I will win?
AI may identify relevant factors, but it cannot reliably predict witness credibility, judicial findings, undisclosed evidence or the way an opposing party will conduct the case.
Does AI replace legal advice?
No. It is best used as a research and organisation tool, subject to professional verification and judgment.
Speak to a Singapore Litigation Lawyer
AI may help you understand a dispute, but significant legal decisions should be based on verified law, admissible evidence and a clear litigation strategy.
Justin Chan Chambers LLP advises and represents clients in:
- civil and commercial litigation;
- family and divorce proceedings;
- cross-border disputes;
- trusts and equitable claims;
- contract disputes;
- building and construction matters;
- MCST and strata title proceedings;
- probate and estate disputes; and
- contested court applications.
Contact Justin Chan Chambers LLP to arrange a consultation if you have used AI to assess your legal position and require a lawyer to verify the advice, review your documents or develop a litigation strategy.
This article is provided for general information only and does not constitute legal advice. AI-generated information should be independently checked before it is relied upon in legal proceedings.