{{item.videoDuration}}
{{item.title}}
{{item.text}}
{{item.videoDuration}}
{{item.text}}
Key takeaways:
- Firms that don’t use AI themselves can’t credibly assess or capture AI value in target companies.
- AI shifts deal origination from efficiency to alpha—surfacing non-obvious targets before they reach competitive processes.
- Institutional memory built on AI compounds with every deal, giving firms a proprietary and appreciating analytical edge.
- Real AI infrastructure requires traceability, codified workflows, and systematic validation—not just tool licenses.
- Use case clarity, capacity to learn, and incentive alignment determine whether a firm’s AI investment delivers results.
How PE firms and corporate acquirers can embed AI into deal origination, diligence, underwriting, and execution
Dealmakers evaluating targets on AI readiness often face an uncomfortable question: Have you done it yourself?
In our last article, we introduced an evaluation framework for assessing whether AI accelerates or erodes value in a target company. The decisive point in the framework is execution capability—whether the organization can absorb AI at the pace required. But assessing execution capability in target companies demands firsthand experience building it. The firms that embed AI into their own deal process can gain both credibility and two compounding advantages: better deal execution today and an institutional asset that sharpens with each transaction.
The owner constraint: A firm’s lack of AI capabilities can constrain their portfolio companies
Whoever acquires a company sets the ceiling on how much AI value it can create. If you diligence AI readiness at the asset level but lack those skills within your own organization, you'll either fail to recognize the opportunity or underwrite value your firm isn't equipped to deliver.
This matters now more than it did a year ago. Given the pace of change in AI, there's no substitute for using it yourself to develop an intuition for what's achievable and the change management processes needed to sustain it. Deal teams that embed AI into their own processes—sourcing, diligence, integration planning and ongoing monitoring—build a fluency that sharpens their ability to imagine what's possible at the target level. That fluency becomes conviction: which opportunities to pursue, how aggressively to underwrite them, and where AI-driven value creation is real versus aspirational. The same holds for corporate acquirers who will recognize integration paths a less experienced buyer is likely to miss.
Many acquirers apply an AI lens to targets while their own deal teams operate with yesterday's tools. The gap between what they evaluate and what they practice is a ceiling on deal performance. For PE firms, it’s a ceiling that will increasingly show up in fundraising too, as LPs begin asking whether a team's capabilities match the value creation story it's selling. We expect LP diligence will shift from "do you have AI tools?" to "show us how AI informed a specific decision and what the outcome was." For GPs whose thesis depends on AI-driven value creation, the ability to point to their own AI-enabled workflows, and connect them to better sourcing or faster value identification, is becoming part of how their track record gets substantiated.
From efficiency to alpha: shifting from a cost savings mindset to a value generating mindset
Many firms default to a cost savings mindset when thinking about AI in their own operations. Compress timelines. Reduce headcount. Do the same work faster and cheaper, focusing on cost savings while the platform beneath you shifts. Efficiency gains are now table stakes. The real opportunity is alpha: using AI to see what others miss, preempt deals before competitive bids, and identify value creation pathways that weren’t visible before.
We have seen these benefits in several ways.
The firms that treat AI as a growth lever for their own capabilities—not just a cost lever—will separate. The efficiency gains come as a byproduct. But the strategic value in these levers is the expansion of what’s possible: compressed value creation timelines, faster paths to EBITDA expansion, and the ability to navigate risk on assets that would otherwise fly under the radar.
Building an AI-centered institutional memory that compounds
Knowledge flows poorly in most deal organizations. Imagine a firm that passed on a target three years ago. The reasons for passing live in an email thread, a marked-up CIM, and a brief committee discussion. Now the company is back on the market, with a different deal team in the lead. Any prior intelligence is effectively lost.
Leading firms are building something different: permanent knowledge bases that capture prior assessments, what worked and what didn’t, which risks materialized—and make it all queryable across the entire deal history. When a deal team evaluates a target, they can leverage AI to surface patterns. How did similar revenue profile companies perform post-close? What red flags were present in similar prior deals? Which assumptions held and which didn’t? Each deal adds to the body of information, or the “corpus.” The AI’s pattern recognition improves. In turn, the firm gets smarter in a way that doesn’t depend solely on people’s memory.
Some firms are going further: reverse-engineering their own investment memos to encode their decision-making framework into AI. The questions the committee always asks. The risks the firm weights most heavily. The patterns that predicted success or failure. The result is AI calibrated to the firm’s investment philosophy—not generic analysis, but judgment shaped by every deal the firm has done. That’s institutional IP, and it appreciates with use.
Moving from AI for AI’s sake to true AI infrastructure
Not every AI initiative delivers. Plenty of firms have purchased licenses, run pilots, and announced AI strategies without meaningfully changing how deals get done. The difference we've seen between building real capability and AI for the sake of AI is building true infrastructure. There are three areas where we've seen companies find success in making their AI programs last:
The people framework begins at home—with the firm
The three levers from Part 1—use case clarity, capacity to learn, and incentive alignment—apply just as directly within the firm’s own deal organization. They determine whether AI becomes part of how deals get done or if it remains a set of isolated tools and pilots.
And as with portfolio companies, this is fundamentally a leadership challenge. Without senior leaders modeling AI usage, creating space for learning, and rewarding new ways of working, the tools sit unused, the pilots don’t scale, and an AI implementation never reaches its full potential. And firms will likely continue to struggle to help their portcos build something they can’t fully achieve.
The irony is sharp. Firms that can’t solve these problems internally will struggle to credibly assess execution capability in targets as we discussed in Part 2. And they’ll struggle to help portfolio companies build it post-close. You can’t pull levers you haven’t learned to operate yourself.
Our PwC Deals perspective
Building AI capability in your own deal process isn’t optional—it’s the foundation for credibility and performance across the deal lifecycle. You can’t assess execution capability in targets if you haven’t built it yourself. You can’t help portfolio companies solve the people problem if you don't have experience with solving it yourself.
To start, PE firms should identify where they are on the AI maturity curve.
- Stage 1 — Experimentation. Licenses bought, individual use, no owner.
Next move: Pick two or three workflows to focus on, make AI the default in them, and assign an owner. - Stage 2 — Workflow integration. AI embedded in specific parts of the process (diligence, document review, sourcing screens). The work is faster but not yet smarter.
Next move: Start building the corpus and stand up data governance before it matters. - Stage 3 — Institutional corpus. Private queryable knowledge base of prior deals. Codified AI agents for recurring workflows. Outputs traceable, systematically evaluated.
Next move: Operationalize across the firm, build the LP-facing story, and extend into portco support.
PwC Deals has embedded AI into deal execution—workflows that analyze vast document sets and surface findings with traceable sources, executed thousands of times across deals globally. We’re building the corpus with every engagement: training our AI on proprietary deal data, refining extraction accuracy, and running systematic evaluations that measure output quality against human-reviewed findings.
Each deal makes the system sharper. The firms that build these types of capabilities will be better positioned to create value post-close—and to define the value creation roadmaps that make AI transformation concrete and accountable. But building AI capability in portfolio companies requires a different playbook. Part 4 will address that challenge.
{{filterContent.facetedTitle}}
{{contentList.loadingText}}
{{contentList.loadingText}}
Follow us
