Skip to main content

The Banker’s Conversation

Lawyering in the Age of AI

Artificial Intelligence (“AI”) is no longer a distant concept reserved for technology companies and research laboratories. It is already reshaping professional work across sectors, including healthcare, finance, education and manufacturing. The legal profession is no exception.

Across the world, lawyers are increasingly using AI-powered tools to conduct legal research, review documents, draft contracts, summarise complex materials and manage large volumes of information more efficiently. These developments create significant opportunities for legal practitioners, but they also introduce risks that require careful professional judgement, disciplined oversight and clear governance. Recent developments across several jurisdictions demonstrate both the promise and the dangers of AI in legal practice.

When Artificial Intelligence Goes Wrong

A recent United States federal case, Withers v City of Aberdeen, highlighted a growing concern within the legal profession. In that matter, a Mississippi federal judge sanctioned lawyers on both sides after court filings were found to contain fictitious legal authorities generated through unverified AI-assisted research.

The court concluded that the lawyers had relied on AI-generated legal research without independently verifying whether the cited cases and authorities actually existed. The consequences were serious: lawyers were fined, removed from the matter, and certain practitioners were barred from appearing before that court for two years.

Although the facts arose in a foreign jurisdiction, the lesson is universal: AI can be a powerful assistant, but it is not a substitute for professional judgement.

South African courts have reached similar conclusions. In Parker v Forsyth NNO and Others, lawyers were criticised after relying on fictitious case references generated through ChatGPT in a defamation-related dispute. The court imposed a punitive costs order and emphasised that modern technology must still be accompanied by independent reading and verification.

More recently, in Mavundla v MEC: Department of Co-Operative Governance and Traditional Affairs KwaZulu-Natal and Others, the High Court criticised legal representatives for citing authorities that did not exist, describing the conduct as irresponsible and unprofessional and referring the matter to the Legal Practice Council. In Northbound Processing (Pty) Ltd v South African Diamond and Precious Metals Regulator, the Gauteng High Court again confronted fictitious AI-generated authorities and referred the legal team to the Legal Practice Council. These South African decisions are particularly relevant to Botswana because they reflect common-law courts in the region applying familiar professional duties of diligence, candour and verification to AI-assisted legal work.

Why Human Judgement Still Matters

AI can assist lawyers, but it cannot replace them. These tools can support legal research, summarisation, document review and the preparation of first drafts. However, AI systems can also produce inaccurate, incomplete or entirely fabricated information.

The danger is that incorrect information is often presented confidently and persuasively, making errors difficult to detect without careful review. For lawyers, every AI-generated output must therefore be independently verified before it is relied upon or submitted to a client, regulator or court. The central principle emerging from recent judicial decisions is that professional responsibility cannot be delegated to technology.

A lawyer who signs a pleading, opinion, contract or court filing remains accountable for its contents, whether the work was prepared personally, by a junior practitioner or with the assistance of an AI tool. This principle aligns closely with the professional obligations that already govern legal practice in Botswana. Duties of competence, diligence and integrity toward the courts do not diminish merely because technology has become part of the drafting process.

How AI is Transforming Legal Services

Notwithstanding these risks, AI presents substantial opportunities for law firms. Tasks that previously required hours of manual effort can now be completed far more efficiently, including:

Legal research and information retrieval;

Contract review and due diligence exercises;

 Analysis of large volumes of documents;

 Drafting of preliminary agreements, pleadings and opinions; and

 Summarisation of lengthy reports and legal materials.

The legal profession is therefore more likely to evolve than disappear. Properly governed, AI will augment lawyers rather than replace them.

Governance, Confidentiality and Data Protection

As AI adoption increases, law firms must ensure that innovation remains balanced against professional obligation. This requires rigorous verification of AI-generated outputs, robust protection of confidential client information, appropriate data security and privacy safeguards, ongoing staff training and clear internal policies governing the use of AI tools.

For Botswana firms, these considerations intersect directly with obligations under the Data Protection Act, 2024, professional conduct requirements and duties of confidentiality owed to clients. The Act commenced on 14 January 2025 and strengthens obligations relating to lawful processing, accountability, data security, breach response, cross-border transfers and data subject rights.Clients are also becoming more informed and increasingly ask how AI is being used on their matters, and what safeguards exist to protect their information.

The Competitive Advantage of the Future

The real competitive advantage will not lie in access to AI alone. Over time, AI tools will become widely available. What will distinguish successful firms is their ability to use AI responsibly, effectively and ethically.

Firms that develop strong governance frameworks, invest in training and maintain rigorous quality-control processes are likely to gain significant advantages in efficiency, responsiveness and client service. For law firms in Botswana, the question is therefore no longer whether AI should be used, but how it should be integrated into practice responsibly.

Recent Research and Strategic Insights

Recent empirical research confirms that the risks associated with AI in legal practice are not merely theoretical. A 2025 Stanford evaluation of leading AI legal research tools found that even systems designed specifically for legal research can produce incorrect or unsupported answers, with hallucination rates reported between 17% and 33% in certain tools. Earlier research on general-purpose large language models identified even higher rates of legal error. The practical lesson is clear: using a specialised legal AI platform may reduce risk, but it does not eliminate the lawyer’s duty to verify every authority, quotation, factual assertion and legal conclusion.

At the same time, global adoption is accelerating. Recent professional-services research indicates that legal professionals increasingly view generative AI as a productivity tool, particularly for saving time, improving workflows and strengthening client service. This creates a strategic tension for law firms: moving too slowly may affect competitiveness, while moving too quickly without controls may expose the firm to professional, regulatory and reputational risk. The appropriate response is neither resistance nor blind adoption, but disciplined integration.

International and regional court developments also show that judges are increasingly treating AI misuse as a matter of professional responsibility. Courts in the United States and South Africa have warned, sanctioned or referred lawyers where filings contained fabricated authorities or where lawyers failed to supervise AI-generated work. These developments are significant for Botswana because they demonstrate how traditional duties of competence, candour, diligence and accountability are being applied to new technology without waiting for entirely new rules.

The Botswana context makes these issues especially important. Botswana is advancing its digital transformation agenda, while international assessments note growing interest in AI, stronger mobile connectivity and increasing attention to digital governance. The Data Protection Act, 2024, which commenced on 14 January 2025, also creates a more demanding compliance environment.In practice, any use of AI involving client information must now be assessed through the lenses of lawful processing, data minimisation, confidentiality, security safeguards, cross-border transfers, breach readiness and accountability.

A particularly important insight for Botswana law firms is that AI governance should not be treated as an information-technology issue alone. It is a leadership, risk-management and client-trust issue. Firms should consider adopting internal AI policies, approved-use lists, matter-level risk assessments, verification checklists, confidentiality protocols, staff training and clear rules on when client consent or disclosure may be appropriate. These measures can help firms benefit from AI while preserving the professional standards that clients and courts expect.

Conclusion

Artificial Intelligence is poised to reshape the legal profession by improving efficiency, reducing administrative burdens and enabling lawyers to deliver services more effectively. However, AI is not a replacement for legal expertise, professional judgement or ethical responsibility.

The future belongs not to AI alone, but to lawyers who can combine technological capability with sound legal judgement, professional integrity and client trust. Human judgement therefore still wins especially when AI does the heavy lifting. 

Disclaimer

This article is provided for general information purposes only and does not constitute legal advice. Readers should seek appropriate professional advice before acting on any information contained in this article.

Authors: 

Lebogang George (Partner – Akheel Jinabhai & Associates in Association with McKee Commercial Law )

Reatile Balopi (Pupil Attorney – Corporate Commercial Department)