Artificial Intelligence in Private Equity


Private Equity
Apr 28, 2023
Artificial Intelligence – AI – is the hot topic of the moment and understandably so, with the widespread availability of services such as Chat GPT bringing the subject firmly into the mainstream and making it widely apparent that we are on the cusp of a major technological step forward.

The general response to AI seems to be a combination of fascination and fear. Fascination both with the concept itself and “what can it do for me?”, fear over “what might it do to me?”. There has been much theorising over how machines driven by AI might develop and how they could impact humankind – I don’t intend to join that debate as, interesting though it is, I find it more helpful to focus on the more immediate practical implications.

And practical implications there most certainly will be, particularly in the field of finance. For the purpose of this post, I will focus specifically on how I see it impacting on the field of Private Equity and what that might mean for careers in the sector.

Before we look into what AI can do in PE, we need to consider the core components of PE investing. In a nutshell, it involves evaluating many potential investments (often without the benefit of an established track record), deciding which are the most promising, doing in depth research to evaluate their potential (based on the business model, the operating environment and the management), working out how the PE firm can add value and eventually exit the investment at a suitable profit.

How much of this procedure could, at least in theory, be replaced by AI (in other words, which human jobs could end up being replaced by machines)? This is where we risk getting drawn into a discussion of what is possible now and what might be possible in the future. It is worth remembering that AI is based on algorithms which are designed to perform a certain task; while self-improving algorithms may widen the scope to automate over time, it seems unlikely that there will be any immediate ‘off-the-shelf’ packages that can contribute significantly to most of these procedures. If correct, this means that PE firms will need to develop tailor-made packages to do any significant tasks. My belief is that those firms that can afford it will do so on an experimental basis, with a view to rolling out the technology, initially in a largely supportive role, to the more mundane tasks once they have established that it can perform these reliably. While firms’ public pronouncements may sound more adventurous, I believe that it is unlikely that they will risk their performance by delegating it to machines until they are confident that it will work.

What might these tasks be? The areas that would seem a good place to deploy early-stage AI include deal sourcing and market intelligence, using AI to scan and analyse huge amounts of data to help spot potential opportunities, as well as analysing those opportunities individually. Similarly, large scale data analysis could help with risk management, spotting things that humans might miss. AI could also assist in the due diligence process, monitor the performance of investee companies and even identify areas for improvement.

Other tasks will, I believe, remain the domain of humans. A large part of PE work is based upon making qualitative (as opposed to quantitative) judgements. As long as investee companies themselves are run by humans, human interaction and relationship building will remain a key component of successful PE investing. While machines may learn to mimic human behaviour, they will never (in my opinion) become sentient beings. The task of building relationships based on trust, confidence, negotiations and personal dynamics will remain a core requirement that cannot be machine replicated. Ethical and moral considerations, and possibly also regulatory requirements, will require human input. Furthermore, AI is trained on historic data and is unlikely to have the human ability to adapt to unforeseen events in the same way that humans can.

What does this mean for careers in the sector? In the very short-term, probably not that much, although there may be more demand for sophisticated coders who are able to develop tailored AI applications for PE firms. As these applications become more embedded, I would envisage them becoming a support tool for (human) executives, helping them to monitor and analyse greater volumes of data more thoroughly and accurately. At this stage, those firms that have developed genuinely value-adding applications will start to gain an advantage over those that haven’t. As and when AI starts to replace humans, I would expect it to be junior analyst and admin roles that will suffer first (this could pose a longer-term problem for the industry – if there is no entry-level mechanism, there risks being a shortage of experienced executives further down the line). Beyond that, it is possible to envisage a future where humans play little or no role PE investment, although personally I think this is unlikely; if you want to be pessimistic, yesterday’s hot topic – environmental collapse – will probably get there first. If, like me, you are more optimistic, technology will play a growing role in supporting human endeavour but humans themselves will remain the driving force.