The first thing we do, is replace all the lawyers with AI
The effects of AI on lawyers are likely to be rather less disruptive than many expect
It has by now become common wisdom that AI will disrupt the legal industry. And when something becomes common wisdom, my spidey sense starts acting up. So much is unknown about the effects of AI on white collar knowledge work, that I am skeptical of any claim made with certainty about what those effects will be.
I suspect that AI tech will complement most knowledge work, rather than disrupt it. I think that AI tech is generally a force multiplier, in that it makes people who know how to use it more productive. For example, consider this study, conducted by Wharton, Harvard Business School, and Boston Consulting Group, which found that consultants who use generative AI tools like ChatGPT are more productive than those who don’t. You can read their actual research here. You can also read Wharton professor Ethan Mollick’s conclusions here. Quoting from Mollick:
In line with our theories, and as we have discussed, we found that the consultants with AI access did significantly beter, whether we briefly introduced them to AI first…or did not. This was true for every measurement, whether the time it took to complete tasks, the number of tasks completed overall…or the quality of the outputs. We rated that quality using both human and AI graders, who agreed with each other…
So, given studies like this, I’m not yet convinced that AI will disrupt all the things that people say it will disrupt. I am not convinced of this because humans are more resilient and adaptable than we often give them credit for.
Let’s consider the arguments in favor of, and opposed to, the proposition that AI will disrupt the legal industry. The point of this exercise is not to pick on lawyers, per se, but rather to understand the kinds of arguments offered in favor of, and against, the proposition that AI will disrupt knowledge work. We could come up with similar arguments in favor of, and against, the proposition that AI will disrupt programmers, management consultants, financial analysts, etc.
Arguments in favor of the proposition.
Efficiency and Speed: AI can process and analzye vast amounts of legal documents faster than humans, enhancing productivity and reducing time on routine tasks.
Cost Reduction: Automation of routine tasks can significantly reduce labor costs, making legal services more accessible.
Accuracy and Consistency: AI can offer consistent and error-free analysis, especially in document review, reducing the risk of human error.
Predictive Analysis: AI can predict legal outcomes based on historical data, aiding lawyers in strategy formulation.
Access to Justice: AI can democratize legal advice, making basic legal services available to those who cannot afford traditional legal fees.
Innovation in Legal Research: AI’s ability to rapidly analyze case law and statutes can lead to more comprehensive and efficient legal research.
Customized Legal Solutions: Machine learning algorithms can tailor legal solutions to individual cases, enhancing client satisfaction.
Emerging Legal Tech Market: The growth in legal technology startups indicates a market shift towards embracing AI in legal services.
Arguments against the proposition
Ethical and Privacy Concerns: The use of AI raises questions about data privacy, client confidentiality, and ethical decision-making in law.
Complexity of Legal Reasoning: AI may struggle with the nuances and complexities of legal reasoning that require human judgment and empathy.
Lack of Trust and Reliability: Clients and lawyers may distrust AI decisions due to lack of transparency in how AI arrives at conclusions.
Regulatory and Legal Hurdles: The legal industry is heavily regulated, and integrating AI poses significant legal and compliance challenges.
Unemployment Concerns: AI could displace jobs, particularly for paralegals and junior lawyers, leading to resistance within the profession.
Quality of AI Training Data: Biases in training data can lead to biased AI decisions, potentially perpetuating injustices in legal outcomes.
Cost and Complexity of Implementation: Small firms may find the cost and complexity of implementing AI prohibitive, leading to a digital divide in the legal industry.
Overreliance on Technology: Overdependence on AI could erode critical thinking and professional skills of lawyers.
Rather than think in terms of disruption, what if we think in terms of how AI can complement lawyers’ work? What do those arguments look like?
Automated Document Review and Management: AI can quickly sift through and organize vast amounts of legal docuemnts, identifying relevant case law, statutes, and legal precedents. This allows lawyers to focus on the more nuanced aspects of case preparation and strategy.
Enhanced Legal Research: AI tools can conduct comprehensive legal research much faster than a human. They can track down case laws, statutes, legal articles, and other relevant documents, providing lawyers with a solid foundation to build their cases.
Predictive Analysis: AI can analyze historical data to predict outcomes of legal proceedings. This doesn’t replace a lawyer’s judgment but provides additional insights for better-informed decision-making.
Drafting Assistance: AI can assist in drafting legal documents by suggesting language, clasues, and legal structures based on vast databases of legal texts. This speeds up the drafting process and helps ensure that documents comply with current laws and regulations.
Routine Task Automation: Tasks like billing, scheduling, and client communication can be automated using AI, freeing up more time for lawyers to focus on complex legal work.
Client Relationship Management: AI tools can help manage and analyze client interactions, providing insights into client needs and preferences, which can improve client service and satisfaction.
Customized Legal Advice: AI algorithms can help tailor legal advice to individual clients’ circumstances, ensuring that the guidance is highly relevant and specific.
Continuous Learning and Development: AI systems can keep lawyers updated on the latest legal developments, new case law, and emerging trends in their field of practice.
Decision Support Systems: AI can act as a decision support tool, offering various perspectives and considerations in complex legal matters, aiding in more comprehensive and well-rounded decision-making.
Enhanced Access to Legal Services: AI can help provide basic legal services to underserved populations, expanding the reach of legal advice and support.
Risk Assessment: AI can assess risks in contracts, mergers, acquisitons, and other legal agreements, providing lawyres with detailed risk analyses.
Legal Analytics: AI can analyze court decisions, judge behaviors, and litigation trends to offer strategic insights that can be used in litigation and negotiation.
I think this provides more than enough evidence to conclude that the world is complex, and that the outcome of AI infiltrating knowledge work is not as clear cut as is sometimes claimed. AI technology is real, and it is here. Its capabilities and its effects are often overhyped, especially by technologists who don’t have any experience operating in the industries which they say it will disrupt. But the other extreme is equally invalid: sticking your head in the sand and pretending that AI doesn’t exist, or claiming that it is all typical Silicon Valley hype, are equally myopic. The best posture is to have your head on a swivel, paying attention to the technology as it develops, understanding its effects as it suffuses and pervades the work you do, and adapting as appropriate.