AI's potential threat to portfolio company value is, in some cases, severe enough to kill deals. That's according to Tobias Haefele, founder of AI advisory firm Riplo and former AI adviser to Hg Capital, who tells Side Letter that at least one fund client has "walked away" from a deal because they weren't comfortable following an analysis of its potential to be disrupted.

Launched last year, Riplo helps private equity firms understand how AI could impact portfolio companies either pre-acquisition or pre-exit. It completed a £2.3 million ($3.1 million; €2.7 million) pre-seed funding round in June led by Cherry Ventures with participation from Blue Lion Capital, founders of QuantumBlack, and angels from McKinsey, OpenAI and Hg Capital, per a statement.

"What often happens is that the IC has a very vague concern," explains Haefele. "We identify different competitive scenarios and [the] breakdown ultimately is, who would actually disrupt you – is this a start-up? Is this an existing competitor? Is this an adjacent entry? Or is that a big tech firm?" For each of these scenarios, Riplo then extrapolates the current ‘tech trajectory’ forward and offers a view on how it will affect the firm in question.

Riplo's role in assessing the viability of assets within the context of AI can create tensions depending on which part of the firm has hired them. "There is definitely a tension depending on who you work for," Haefele says. "If you work for the IC versus… the deal team, then of course the deal team might… want to feel like they're making an informed decision, but once they have made up their mind, they would want the diligence to help them tell their story as opposed to kill a deal. And so there's a balance to be struck in terms of how, as a diligence vendor, you position yourselves there."

The role is a crucial one, however, given industry concerns over the erosion of equity value. "Irrespective of whether we kill a deal or not, what we've seen happen… is either that the IC shoots down a deal that was somewhat unexpected because of AI, or that the lenders shoot down a deal, which the PE funds are not used to," says Haefele.

"The providers of debt say, 'Actually, we don't want to provide debt on this because we don't feel that you guys understand the AI risk well enough.' And this is part of the reason why the funds want, even just psychologically but [also] conceptually, a separate piece that explicitly speaks to whatever concerns you might have around AI on a certain deal."

These considerations also come into play at the time of exit. "This is almost more interesting because then we can give them a view of what we would say if we were on the buyside," Haefele adds. "There's also an element where the funds think… 'Do we want to continue holding this vehicle, or how should we think about what happens next? Should we sell out to our own funds, or keep some form of exposure to it?'"

AI has, so far, raised more questions than answers for many LPs assessing the potential impact (both good and bad) on their portfolios. Entities capable of providing even a shred of clarity around the situation may find themselves in high demand.