Around 85 per cent of the jobs that will exist in 2030 haven't been invented yet. This was the claim in 2017 from a report by the Institute For The Future. Whilst it's still uncertain if those numbers will stack up, we're certainly seeing an explosion of new roles created by our changing environment and evolving technologies. Artificial intelligence is one of these technologies and although AI has been accused of making swathes of the workforce redundant, this type of technology is creating opportunities and new roles, although how long these roles will be relevant in this fast-moving world is up for debate.
One of these new roles is prompt engineer – an individual who has a deep understanding of AI, and particularly the Large Language Models (LLM) wrappers like ChatGPT (based on the OpenAI engine). These prompt engineers can redefine the questions posed to an AI or LLM system to get optimal results. In these instances, 'optimal results' can refer to getting the best answers, as well as getting these answers efficiently, with a minimal amount of back-and-forth prompting.
However, AI models are rapidly evolving. Today they require less prompting and will continue to need less input over time, and importantly, less prompting indicates the AI is learning faster. As a result of these factors, and despite the recent hype around this new job opportunity, some believe the need for prompt engineering is not as significant as first understood. Instead, the real area of need is problem formulation. This skill involves being able to properly diagnose and define a problem, so the AI model knows what it needs to work on. And it's something lots of people are bad at – a survey revealed that 85% of C-suite executives agreed that their organisations were poor at problem diagnosis, for example. Countless leadership experts agree that the old adage 'bring me solutions, not problems' is an unhelpful approach, leading to a failure to properly define and interrogate a problem before a solution is decided, yet the thinking persists.
There's no doubt that problem formulation will be key to the successful adoption and use of AI for corporate legal teams and law firms but it's also good practice more generally for leaders who want to effect change and improve efficiencies. So what is it?
At its core, problem formulation first requires a diagnosis of the issue and by getting to the heart of the problem, the key objective can be defined. Once this is recognised, it's important to break the problem down into digestible portions and these can be related to the people, process and protocols involved. Next, it's about reframing the problem. Thinking about the issue from multiple perspectives and how the problem impacts different stakeholders can be hugely helpful. For example, a certain process might be causing frustrations internally due to its inefficiencies, but thinking about how this process also impacts a client, or other members of the team, can help you reexamine the issue and as a result, arrive at a more creative and holistic solution.
Finally, including boundaries to a problem can also be really helpful when using AI tools to find a solution. For example, interweaving your tone of voice or brand guidelines if you need text generating can ensure the output is consistent . Conversely, removing boundaries or creating new ones can help give a fresh perspective if that's what is required.
There's no doubt that AI tools, particularly those like ChatGPT which are based on language models, are an exciting area that create opportunities for innovation. The most significant short-term opportunities presented by this technology will likely be optimising internal processes and it could also be used for creating simple templates, contract management, administrative automation, and some document review. All need a full understanding of the issues that need fixing to begin with – problem formulation. At Epiq we're focused on how these tools can be utilised for both corporate legal departments and law firms. Reminders around security, compliance, and information summarisation, for example, across commercial contracts are all use cases we have identified. We continue to look for new ways to incorporate AI into our Discovery solutions and interactions, including Epiq Access, which provides our clients with intuitive and interactive dashboards and reports across their legal team.
There's a lot of hype surrounding AI and Generative AI which often needs an expert eye to interpret – legal teams don't need a prompt engineer, but a serious look at problem formulation is definitely an advantage – and not just for utilizing AI tools, but improving efficiencies across the business.
Authors:
Shah Karim is the Chief Technology Officer for Legal Solutions at Epiq. In this role, Shah heads R&D and technology strategy for Epiq's global legal solutions business, working closely with our clients and partners to deliver best in class technology that supports solutions.
Eric Crawley is a managing director of document review services for Epiq. His responsibilities include leadership and oversight of review management and technology assisted review consulting.
