The AI valuation shift: a point of view for industrial CEOs and their deal teams

Industrial M&A has long operated under a framework anchored in tangible assets and predictable cash flows: EBITDA multiples benchmarked to peers, order backlogs, capacity utilization, customer concentration and asset replacement cost. Discounted cash flow models project revenue growth based on end-market demand, pricing power and capital expenditure cycles. These frameworks assumed competitive advantage was rooted in physical scale, installed base and supply chain reach.

That playbook contains structural blind spots exposed by AI:

1. It undervalues intangible assets — decades of proprietary operational data in plant historians and sensor logs do not show up on the balance-sheet — yet may be the most strategically valuable asset a target possesses.
2. It struggles with non-linear value creation: as one example, a predictive maintenance model trained on five years of vibration data compounds in accuracy with every cycle, creating accelerating returns that a linear discounted cash flow (DCF) cannot capture.
3. It applies uniform discount rates — failing to distinguish between companies whose operational risks are actively reduced by AI and those relying on manual processes that introduce variability and human error.

The EY Deal Barometer captures the resulting dynamic: US deal value surged 36% in 2025, with billion-dollar-plus transactions rising to 27% of all deal activity — up from a 22% pre-pandemic average. Corporate deal decisions are now primarily growth-driven and tied to transformation roadmaps, with transactions most active in HVAC, air purification, battery storage and life sciences-adjacent manufacturing — verticals where AI-enabled monitoring, optimization and outcome-based service models are most advanced.

1. Predictive operations as margin expansion

The U.S. Department of Energy’s Federal Energy Management Program documents that a properly functioning predictive maintenance program delivers a 25% to 30% reduction in maintenance costs, 70% to 75% fewer breakdowns, 35% to 45% less downtime, and a tenfold ROI. Predictive maintenance targeting energy efficiency yields up to 20% annual energy savings.

Industry analysis shows that unplanned downtime now costs the Fortune Global 500 roughly $1.4 trillion each year — about 11% of total revenues — with automotive plants incurring losses of up to $2.3 million per hour. Major manufacturers suffer an average of 25 downtime incidents per month per facility. While average downtime hours have declined by nearly a third since 2019, per-hour costs have risen dramatically due to inflation, energy prices and supply chain complexity.

The EBITDA impact is easy to translate. A facility with 400 hours of annual unplanned downtime at $6,700 per hour loses $2.7 million annually. A 35% to 45% reduction saves $950,000 to $1.2 million per plant — before maintenance cost savings and equipment life extension. Leading manufacturers are already operating at this scale: a major automaker saved $20 million annually deploying AI monitoring on assembly robots. A major automotive OEM predicted 22% of component failures 10 days in advance, saving 122,000 hours of downtime and $7 million on a single component type. For acquirers, this is all EBITDA and value creation opportunity — not a technology story.

2. AI-enabled products and revenue durability

AI is enabling industrials to shift from selling equipment to selling outcomes, fundamentally altering revenue profiles and the multiples they command. Industrial OEMs historically traded at 8x-12x EBITDA — well below the 15x-25x multiples in software and recurring-revenue businesses. AI-powered connectivity is converging and blurring the lines in how these industries are valued.

By embedding sensors, real-time monitoring and AI analytics into products, manufacturers are creating outcome-based revenue streams. Instead of selling an HVAC system, a manufacturer sells “comfort-as-a-service.” Instead of selling a water pump, an industrial company will sell volume moved. Instead of selling a compressor, they will sell guaranteed uptime per hour. Top-quartile manufacturers now derive up to 40% of sales from innovative products, services and software. IoT Analytics reports 63% of industrial providers with recurring models achieved 29% CAGR in service income over three years.

The EY CEO Outlook confirms the demand side: 49% of US CEOs cite accelerated AI adoption as the single biggest growth factor for 2026. Industrial targets that have converted AI capability into recurring revenue are precisely the assets these CEOs are pursuing — and the blended multiples will reflect it. As more manufacturers embed AI-powered monitoring into their installed base, the line between “industrial company” and “industrial technology company” is blurring — and the M&A market is beginning to price that convergence.

3. Supply chain intelligence as risk reduction

AI-driven demand sensing, supplier risk scoring and logistics optimization systematically reduce earnings volatility — which directly lowers the discount rate acquirers apply. Machine learning algorithms analyze historical sales data, macroeconomic indicators, weather patterns and real-time consumption signals to improve forecast accuracy. Supplier risk models continuously monitor financial health, geopolitical exposure and quality metrics across the supply base.

The valuation impact is material. The difference between a 10.0% and 8.5% WACC on a $2 billion industrial target translates to $200 million to $300 million in additional enterprise value. With GDP growth forecasted to moderate to 1.7% in 2026 amid tariff uncertainty, companies demonstrating AI-enabled supply chain resilience will command structurally higher valuations. Companies that can improve their supply chain intelligence will reduce the impact of trade disruptions — rather than simply weathering them — and present a fundamentally different risk profile that is directly reflected in deal pricing and financing terms.

4. Data assets as hidden balance-sheet value

Decades of proprietary operational data — vibration signatures, thermal profiles, quality inspection records and field service logs — collected for compliance or operations have never been valued as a strategic asset. That is changing. AI models are only as valuable as the data they train on, and domain-specific industrial data is nearly impossible to replicate. A sensor dataset from 15 years across 200 installations can become a real competitive moat.

Acquirers now evaluate data architecture quality alongside more traditional intangible assets such as patents and customer lists. Clean, cloud-connected data lakes are an asset; fragmented data trapped in siloed historians and disconnected legacy systems is a liability that buyers deduct as post-close remediation cost. For industrial CEOs, the implication is clear: data governance, architecture modernization and OT/IT integration are not infrastructure projects. They are enterprise value creation initiatives that directly affect how a future acquirer will price the business.

5. The AI gap discount

The NAM’s Manufacturing Leadership Council reports that while 51% of manufacturers use AI, 82% lack AI-ready skills and 65% lack AI-ready data. Only 28% describe operations as “smart,” though 76% expect to be there within two years. The Stanford HAI AI Index confirms that only a small single-digit percentage of companies achieve AI maturity at scale. While there are certain companies that are extracting selling, general and administrative (SG&A) savings by leveraging AI use cases (less impactful) or through full-scale reimagination (more impactful), many more have made limited or no progress. Sophisticated acquirers are pricing this aspiration-execution gap — underwriting post-close digital remediation as a direct deduction from enterprise value. The AI gap discount is not theoretical; it is reshaping deal outcomes today.

Traditional diligence remains essential but is no longer sufficient. Smart acquirers now layer five AI-specific lenses onto every industrial transaction:

Data architecture and quality

Can the target’s data fuel AI at scale, or is it trapped in siloed historians and legacy SCADA? The NAM found 65% of manufacturers lack AI-ready data — making architecture a critical differentiator. A target with a clean, cloud-connected data lake across its manufacturing footprint is fundamentally more valuable than one with equivalent physical assets but fragmented data.

AI talent and organizational readiness

Is there genuine capability with dedicated data science resources and executive sponsorship, or a single unscaled pilot in IT? With 82% citing skills as the top barrier, talent depth is a quantifiable integration risk — and its presence is a meaningful competitive advantage.

Model governance and IP

Acquirers should assess ownership of AI models and training data, third-party licensing dependencies, vendor lock-in exposure, and algorithmic bias — particularly explainability for safety-critical and quality-impacting industrial decisions.

Technical debt as hidden liability

Legacy OT/IT stacks that resist AI integration represent quantifiable post-close capex. Aging PLCs, proprietary SCADA and absent edge computing are now diligence line items that sophisticated buyers deduct from valuations.

Regulatory exposure

EU AI Act compliance for industrial automation, export controls on AI-enabled products and evolving workforce monitoring requirements across jurisdictions must be assessed and factored into the post-close investment thesis.

The AI diligence function is itself being transformed by the technology it evaluates. AI-powered tools are enabling due diligence teams to process data rooms in days rather than weeks, identify anomalies with greater precision and model integration scenarios with a level of granularity that manual analysis cannot match.

The EY CEO Outlook found that 97% of CEOs are undergoing or planning enterprise-wide transformation, with AI at the center. The implications are specific and actionable depending on which side of a transaction you occupy.

If you are a potential seller

Translate AI capabilities into valuation language. The difference between “we have an AI initiative” and “AI contributes 250 basis points to our gross margin through predictive operations across 12 facilities” is the difference between a technology story and a valuation driver. Quantify EBITDA impact, document recurring-revenue metrics, prepare a data asset inventory and demonstrate scalability — proven at three plants with a documented playbook for 12 more.

Equally important is articulating the compounding nature of your AI capabilities. Acquirers pay premiums for AI systems that get more accurate with each data cycle, that deepen customer relationships through outcome-based models and that create switching costs through proprietary data and workflow integration. The strongest sell-side narratives frame AI not as a feature set but as a flywheel.

If you are an acquirer

Build internal capability to distinguish AI substance from AI theater. The EY CEO Outlook found 25% of CEOs cite distinguishing hype from commercially viable opportunities as a top hurdle. Conduct technology diligence with industrial data infrastructure practitioners — not just enterprise software generalists — and enter every deal with a clear AI integration thesis.

Equally critical is building a post-merger integration plan that preserves and scales AI capabilities. The fastest way to destroy AI value in a transaction is to integrate the target’s technology stack into a legacy environment that cannot support it. Acquirers should enter every deal with clear answers: which AI capabilities will be preserved, which will be scaled across the combined enterprise, and what investment is required to realize the AI-driven synergies underwritten in the deal model.

If you are on a board

Board members should ask: What is our AI maturity versus peers? What would AI diligence reveal about us today? Are we investing fast enough to avoid the AI gap discount? Eighty percent of manufacturers say AI will be essential by 2030 — the NAM’s Manufacturing Leadership Council data makes clear that the time to close the gap is now, not in the next strategic planning cycle.

Agentic AI

The next generation will autonomously execute multi-step workflows — adjusting process parameters, rerouting schedules and orchestrating maintenance without human intervention. Spending on agentic AI may reach $155 billion by 2030, compounding the operational advantage of early movers and steepening the penalty for those who fall behind.

The autonomous factory

End-to-end AI deployment could drive 30%+ productivity gains in industrial operations. Virtual AI is moving from operator support to autonomous execution; physical AI — including robotic systems guided by AI models — is entering manufacturing environments at scale. China installed 276,300 industrial robots in 2023, over half the global total. The most valuable future assets will be those with digital architecture to support autonomous operations.

New valuation frameworks

“Data-adjusted EBITDA” reflecting AI-driven margin contribution, AI capability scores in confidential information memorandums (CIMs), and multiples formally tiered by AI maturity — analogous to SaaS valuations by net revenue retention — are all emerging. The traditional playbook will not disappear, but it will be augmented by an entirely new analytical layer.

From efficiency to growth

The EY CEO Outlook signals a decisive pivot: 91% of US CEOs see AI as transformative to value creation, not just cost structure. The next cycle’s winning acquirers are not looking for AI that cuts costs — they want AI that creates new revenue categories, deepens customer relationships and builds compounding competitive moats.

The AI valuation shift is not a future scenario to monitor. It is reshaping how industrial deals are sourced, evaluated, priced and integrated today. For industrial CEOs, the strategic question is not whether AI will affect enterprise value — it already has — but whether your company will carry a premium or a discount. For the industrial sector specifically, this moment carries particular urgency as the leading industrials pull ahead on AI maturity — the gap between AI-ready and AI-exposed assets will widen, and the market will price it with increasing precision.