A new McKinsey & Company AI report  "The State of AI in 2025: Agents, Innovation, and Transformation" (November 2025) just dropped, and it's telling a very different story than the doom-and-gloom narratives you might have seen circulating about enterprise AI adoption.

While MIT's recent research  reported a 95% failure rate for enterprise AI implementations, the study revealed a critical pattern: external partnerships focused on learning-capable, customized tools reached deployment ~67% of the time, which is twice as likely to succeed as internal builds. This distinction matters.

McKinsey's latest survey of 1,000+ organizations reinforces why:
We're not failing at AI. We're in the experimentation phase. And that's exactly where breakthrough adoption begins.

For private equity and private markets professionals evaluating AI solutions, we break down the key insights from McKinsey's latest report and what they mean for your firm. Transformation is the buzzword, but how quickly your firm moves from curiosity to measurable impact is what counts. 

1. Experimentation is the norm and that's actually good news 

Nearly two-thirds of organizations haven't begun scaling AI enterprise-wide yet. 

This isn't a red flag. It’s the natural adoption curve for transformative technology. Smart firms are piloting selectively, validating impact on specific workflows before committing to massive deployments.

What this means for PE: 

The opportunity isn't just about moving early but selecting the right technology partner who embraces that we're in the adoption phase. Winners will experiment strategically on high-ROI workflows (CIM analysis, deal evaluation, portfolio monitoring) rather than waiting for perfect enterprise strategies.

Metal's pilot-first approach builds trust before scale:

We're built for exactly this phase. Metal's model is simple: test AI on high-value diligence workflows, validate measurable wins, then scale with confidence across your organization. Our goal isn't just experimentation but securing your trust through proven impact before any enterprise-wide rollout. No operational burden. No years of planning.

2. Curiosity in AI agents is high, but trust requires more than hype 

62% of respondents are experimenting with AI agents.

Curiosity alone doesn't drive adoption. Effective agents require accuracy, reliability, and domain context.

What this means for PE:

Generic AI copilots can answer questions, but purpose-built AI agents understand real contexts: CIMs, financial statements, and investment framework, etc. The difference between "interesting demo" and "tool we use daily" comes down to whether the agent actually understands your workflow context and delivers traceable, trustworthy outputs.

Metal as your institutional intelligence platform:

Our agents and custom workflows are purpose-built for PE diligence and investment workflows. They're grounded in structured company data, industry-specific document parsing, and traceable citations. Every insight links back to source documents. Every analysis follows your firm's investment frameworks.

3. Use-case wins are real, but enterprise impact takes time 

64% of respondents report that AI is enabling innovation, and organizations are seeing tangible cost and revenue benefits at the use-case level. However, only 39% report EBIT impact at the enterprise scale.

What this means for PE:

Focus on workflows where AI delivers immediate, measurable value— hours saved, deals evaluated, insights unlocked. These quick wins build momentum and justify broader investment. Enterprise transformation follows; it doesn't need to come first.

Metal closes the gap between early wins and sustained operational impact:

We start with high-ROI use cases that matter today, such as automating CIM analysis, accelerating company research, and streamlining deal evaluation. These aren't theoretical benefits but actual hours saved per deal and deals evaluated per week.  Because Metal's value shows up immediately in daily workflows, firms realize tangible gains without waiting for enterprise-wide AI strategies to mature. 

4. High performers optimize for growth AND efficiency 

80% of organizations set efficiency as an AI objective, but the companies seeing the most value also prioritize growth and innovation.

What this means for PE:

AI shouldn't just make your existing process faster. It should unlock capabilities you couldn't access before. Analyze more opportunities with the same headcount. Run deeper thematic research. Evaluate markets that were previously too resource-intensive. The real advantage comes from doing things differently, not just doing the same things quicker.

Metal is built for both sides of the equation: 

Efficiency wins are immediate from hours saved on CIM parsing, data extraction, and memo creation. But the real advantage is strategic: analyze more opportunities with the same team, run thematic research that was previously too resource-intensive, explore markets you couldn't justify before. 

We drive growth and innovation by turning your firm's deal history into institutional intelligence. Metal's data infrastructure captures your unique investment frameworks and insights, then activates them. Through custom workflows, we enable shared knowledge that compounds, proprietary insights that deepen with every deal, and broader coverage that drives growth.

That's the difference between experimenting with AI and being a high performer.

5. Workflow redesign separates winners from experimenters 

50% of AI high performers intend to use AI to transform their businesses, and most are actively redesigning workflows rather than bolting AI onto existing processes.

What this means for PE:

The firms gaining lasting advantage are rebuilding core workflows to be AI-native. From automated CIM ingestion to real-time company monitoring and collaborative research, this represents a fundamentally better way of working, not just an incremental improvement.

Metal is built to connect, not replace:

We enable AI-native workflows across the investment lifecycle: automated CIM processing, continuous monitoring, chat-based research. Metal modernizes deal workflows by meeting your team where they already work. No tech stack overhaul required.

We integrate directly with DealCloud, SharePoint, Egnyte, Outlook, and your existing systems, functioning as the connective intelligence layer that unifies documents, insights, and investment decisions.

6. The best firms invest in AI AND their people

Respondents vary in their expectations of AI's impact on workforce size: 32% expect decreases, 43% no change, and 13% increases.

What this means for PE: 

Leading firms understand that AI doesn't replace the need to invest in people but amplifies their impact. As one industry leader put it: firms need to "invest in the future leaders of the firm." AI moves teams faster on execution, but investment decisions still require human judgment, experience, and intuition. The competitive edge comes from combining both.

Metal unlocks your team’s full potential:

We're built to handle the manual, repetitive work: document reading, data extraction, initial analysis.

So your professionals can focus on higher-value activities: building relationships, developing investment theses, making critical judgment calls. 

Metal enables your team to evaluate more opportunities, conduct deeper research, and move faster on live deals.

The Window Is Open. But It Won't Stay That Way

The narrative that "enterprise AI is failing" misrepresents where we actually are in the adoption cycle. Organizations aren't failing but learning, testing, and selectively scaling what works.

For private markets, this moment represents opportunity:

1. The technology delivers real value today on core diligence workflows. T

2. The competitive landscape rewards early movers and can build meaningful advantages. 

3. The risk is low enough through pilot-first approaches that there's no reason to wait.

The experimentation phase won't last forever. Right now, it's exactly where the smart money is moving.

Book a demo to see how leading PE firms are generating measurable impact.