Why traditional M&A due diligence systematically underprices AI risk and overpays for AI theater — and the five-dimension scorecard PE firms are quietly building to fix it.
Axel Tombereau, Odyssey
CEO Briefings
Traditional M&A due diligence often fails to address AI maturity, data asset quality, and agent infrastructure readiness. Both buyers and sellers must now adapt to new evaluation standards.
The Signal
Over the past quarter, I reviewed transaction documentation for seven technology acquisitions in Europe and North America. In each case, sellers emphasized AI capabilities as key value drivers. However, buyers lacked a systematic framework to evaluate these claims. As a result, AI maturity was assessed through informal CTO interviews and basic tool checklists, leading to assessments that were often superficial or misleading.
The M&A market has yet to fully recognize AI as both a significant value driver and a material risk factor in enterprise transactions.
The Strategic Read
There are three key blind spots in current due diligence frameworks. First, data moat quality: due diligence often confirms the presence of data but does not assess whether it is proprietary, well-structured, or strategically defensible. Second, model dependency: while AI tools are inventoried, there is little evaluation of reliance on a single vendor, switching costs, or the portability of agent intelligence. Third, workforce AI fluency: assessments focus on the technical team but overlook whether the broader organization can operate, manage, and evolve AI systems.
Sellers are increasingly overstating their AI capabilities, leading to a credibility gap in transactions. Companies with pilot-stage deployments often describe themselves as “AI-powered,” while those using basic Copilot tools claim “agentic AI capabilities.” Without a rigorous assessment framework, buyers cannot distinguish genuine AI maturity from AI theater, where capabilities are exaggerated for marketing purposes without substantive operational depth.
Private equity firms, especially those with technology-focused portfolios, are beginning to develop proprietary AI scorecards to assess portfolio companies and acquisition targets. These scorecards evaluate operational AI deployment, data infrastructure readiness, AI talent depth, competitive defensibility, and governance maturity. They are becoming integral to the investment thesis, not just the due diligence process.
If you are preparing to sell, your AI maturity now directly influences your valuation. If you are preparing to buy, your due diligence framework likely overlooks this critical factor.
The CEO Move
Regardless of whether you are buying or selling, the recommended action is the same: develop an AI scorecard that systematically assesses five dimensions. These include operational deployment (AI systems in production handling real transactions), data infrastructure (quality, accessibility, proprietary value, governance), AI talent (dedicated team, workforce fluency, organizational readiness), competitive defensibility (data moats, switching costs, workflow IP), and governance (agent oversight, risk management, regulatory compliance).
If you plan to pursue a transaction within the next eighteen months, begin building your evidence base now. Compiling production deployment metrics, data asset inventories, and governance documentation requires significant time. CEOs who prepare early will present a stronger case than those who wait until the process begins.
