How private investors are using alternative data and AI to generate alpha

The other day I was at an alternative data and AI breakfast series. At the breakfast were CTOs of hedge funds, alternative data providers, private equity investors, and a number of data infrastructure start-ups.

Alternative data and AI are no longer two parallel conversations. Alternative data hands you new inputs such as satellite imagery, payment transaction flows, hiring trends, ESG signals, and all manner of digital data. AI gives you the ability to process that at scale and at speed, to turn noise into something resembling a competitive signal. Together, they’re producing tools that simply didn’t exist a decade ago.

Hedge funds were first to see this. The average fund spends about $1.6 million a year on alternative data, with the biggest investing $5 million or more and integrating feeds from up to 40 data vendors. Many of those workflows are fully AI-driven: language models digesting earnings calls, machine vision counting cars in car parks, multi-factor models blending alternative and traditional datasets into predictive scores.

Private equity has not been as quick to adapt. Only 27% of firms use alternative data today, although another quarter plan to start soon. Given how much unmined data sits in sourcing, diligence, and portfolio monitoring, the gap is large and the upside for first movers even larger.

A few private investment firms have moved beyond pilot projects to make AI and alternative data an integral part of their investment workflow.

1. Deal Sourcing

AI platforms surface off-market opportunities before the company is even looking. Bain’s 2025 PE Report found AI-driven sourcing tools make private investors 30% more likely to spot them. Deal flow still relies on networks, inbound pitches, and research teams burning the midnight oil. But AI widens the funnel and shortens the cycle.

Investors are now:

• Mining unstructured data from press releases, hiring spikes, patent filings and product launches.

• Detecting thematic shifts by parsing thousands of funding rounds, analyst notes, and news articles in seconds.

• Scoring targets using predictive models trained on their own win/loss history.

According to Blackstone’s news briefs, over 50 data scientists sit inside its investment teams. They draw on information from more than 230 portfolio companies and 12,500 real estate assets to build sourcing tools that reveal patterns no single team could find on its own.

2. Faster, More Accurate Due Diligence

One mid-market buyout shop cut initial diligence from three weeks to four days by automating 70% of document review. McKinsey estimates AI can reduce review times by up to 60% while improving anomaly detection. Diligence has always been a bottleneck. AI compresses the timeline without sacrificing depth.

Firms now use it to:

• Clean and reconcile financials from multiple systems.

• Scan contracts, compliance files, and litigation histories for red flags.

• Assess competitive position using sentiment analysis of reviews, analyst notes, and market chatter.

And here’s an example case from a Blackstone news brief that’s worth noting. During a diligence meeting, a target presented a “defensible data moat” as a core strength. Within hours, Blackstone’s data team replicated the product using public data and large language models, matched its performance, and determined the moat was far shallower than described. This insight materially changed the deal’s risk profile.

3. Portfolio Value Creation and Monitoring

PwC’s 2025 study found AI-enabled monitoring can lift exit multiples by up to a full EBITDA turn. In practice, Blackstone presents examples of how its portfolio companies use AI. One portfolio company increased AI-handled customer service tickets from 10% to 40% in a year, freeing human agents for complex issues and improving satisfaction for both customers and staff. Other rollouts such as labor optimization, opportunity prioritization, AI-generated content  were delivered faster and with less risk thanks to a centralized AI infrastructure.

Post-acquisition, AI is acting as a multiplier for operations, risk management, and growth. Uses include:

• Real-time KPI dashboards integrating ERP, CRM, and accounting data.

• Churn prediction models for subscription businesses.

• Dynamic pricing engines that move with the market.

For now, there remains a small group of private investors who have invested in using AI and alternative data in their investment processes. So that their machines see what they see — and often what they do not.