Artificial intelligence (hereinafter “AI”) is gradually becoming a common tool in legal practice. Lawyers use it to search for information, analyze large volumes of documents, draft legal documents, organize case materials, review arguments, and work with contracts.
However, the use of AI in a lawyer’s professional practice has one important caveat: responsibility for the outcome of the work still rests with the human. The mere fact of using AI does not constitute a disciplinary offense. At the same time, if a lawyer uses incorrect information without proper verification, discloses confidential data, or allows an algorithm to effectively replace their own professional judgment, this may have disciplinary consequences.
How Lawyers Use AI
AI can significantly simplify the performance of routine tasks. For example, a lawyer can use it for an initial analysis of legislation and case law, structuring case materials, summarizing large volumes of documents, or identifying potential weaknesses in a legal position.
Another area is working with text. Generative models can help draft pleadings, attorney’s letters, legal opinions, and correspondence, as well as edit text that has already been prepared.
AI can also be useful when preparing for litigation.
For example, a lawyer might ask the system to pose critical questions regarding their position, simulate the opponent’s arguments, or draw attention to circumstances that may have been overlooked.
However, it is important to understand that AI can be a lawyer’s assistant, but not a replacement. Generative systems are capable of producing convincing but completely incorrect answers. They can fabricate a court ruling, misquote the law, cite a nonexistent provision, or use a version of legislation that is no longer in effect. That is why any legally significant information obtained from AI must be verified by a lawyer using reliable primary sources.
Can the Use of AI Lead to Disciplinary Liability?
Currently, the Law of Ukraine “On the Bar and the Practice of Law” and the Rules of Professional Conduct for Attorneys do not contain a specific prohibition on the use of AI. These regulations were established before the widespread adoption of generative AI systems.
At the same time, this does not mean that the use of AI falls outside the scope of a lawyer’s professional responsibility. The Rules of Professional Conduct establish general requirements for professional practice regardless of the technological tools a lawyer uses.
In particular, the law requires attorneys to have a high level of professional training and knowledge of the law and its application in practice. Therefore, if an attorney submits information generated by AI to a client or the court without verifying it, and that information turns out to be incorrect, the issue may lie not in the use of the technology itself, but in the improper performance of professional duties.
Equally important is the principle of prioritizing the client’s interests. Incorrect legal advice based on an unverified AI response could influence the client’s decision, financial expenses, or the prospects of a legal dispute.
The rules of attorney ethics also require a lawyer to provide the client with objective information about the legal basis for their position, possible consequences, and adverse precedents. Therefore, relying on an AI response as a definitive legal opinion is unacceptable.
Confidentiality—One of the Main Risks
Another serious issue is attorney-client privilege and personal data. To obtain a useful response from AI, a lawyer may upload a contract, case materials, correspondence with the client, or other documents to the system. However, before doing so, it is essential to understand how a specific service processes the information it receives.
The terms of use for various AI platforms vary. Key factors include the type of account, privacy settings, data retention policies, and the terms governing the use of information to improve models. Therefore, mechanically uploading case materials to a public AI service may pose a risk of breaching attorney-client privilege.
Before using AI, a lawyer should assess exactly what information they are providing to the system and whether they have the right to do so. Particular caution is required when handling clients’ personal data, trade secrets, confidential contracts, materials from criminal proceedings, and information that has come to the lawyer’s attention in the course of providing legal assistance.
A lawyer’s use of AI is not, in and of itself, a violation of professional ethics. The problem arises when a lawyer loses control over the output of the AI system. Unverified court decisions, fabricated legal provisions, erroneous legal conclusions, or the transfer of confidential information to a third-party service can pose serious risks to the lawyer and their client.
Therefore, the basic principle for using AI in legal practice is simple: AI can assist a lawyer, but it cannot assume responsibility in the lawyer’s place. Current rules of legal ethics already contain sufficient general requirements against which a lawyer’s conduct can be evaluated when using such technologies.
Do you need legal advice on using AI in your business or legal practice? Contact our law firm—we’ll help you assess the risks and develop a safe approach to using AI.
artificial intelligence / intellectual property / lawyer / responsibility
