Artificial Intelligence in Private Equity

Page 1

Newsletter (23) - Artificial Intelligence in Private Equity

(1) Introduction In today’s newsletter, we examine the growing role of Artificial Intelligence (AI) across the private equity (PE) lifecycle. While financial engineering and operational improvements have long driven value creation in PE, every GP now claims AI is central to their strategy. But how much of this is genuine competitive advantage, and how much is simply table stakes? From sourcing deals and conducting due diligence to transforming portfolio companies and streamlining fundraising, AI is reshaping how PE firms invest, create value and operate. We examine where AI truly adds value to private equity, and where the hype exceeds the results.

(2) AI in Deal Sourcing and Due Diligence Traditionally, PE deal sourcing has been reliant on relationship networks, intermediary referrals and manual data analysis. In parallel, the industry has seen increasing competition and rising dry powder, while leading firms have begun shifting to proprietary AI-powered sourcing engines, which reflects a broader evolution in how deals are identified and pursued. These AI systems allow private equity firms to scan large datasets, detect patterns humans may miss, and rank targets using quantitative metrics and historical deal data. According to the World Economic Forum, AI can identify about 195 targets in the time it takes a junior analyst to evaluate one. Bain argues that the usage of AI in deal sourcing can increase lead quality by 10-15% and reduce acquisition costs by around 20%. Because private company data is scarce, sourcing requires significant manual effort and many attractive targets remain overlooked. AI tools address this inefficiency by systematically increasing market visibility and improving daily workflows. A prominent example of this is Deutsche Beteiligungs AG’s (DBAG) “Nexus” platform, launched in 2025. Before this AI platform, DBAG had used an Excel sheet with 22,000 lines to manually identify opportunities. The platform aggregates over 500,000 data points from various data sources to keep up to date with the development of all opportunities in the German mid-market segment. The data includes financial metrics, shareholders and much more information, enabling a digital assistant that continuously scans the market to identify tailored situations and high-potential targets. This fully automated and customisable platform uses algorithms and LLMs to eliminate a key bottleneck in private equity deal sourcing. However, as AI sourcing tools become widely available through third-party platforms like Grata or Extruct, the question arises whether these tools create differentiation or simply raise the baseline. The advantage may lie less in having the tool and more in the proprietary data that feeds it. Similarly, Blackstone is using an internal platform, applying AI to pipeline screening. Thus, allowing faster processing and better prioritisation of high-potential opportunities. The company uses machine learning to identify early-stage targets before they reach the broader market. In contrast, EQT’s Motherbrain is designed not only as a sourcing engine but also as an institutional knowledge system. It integrates proprietary data such as meeting notes, prior investment evaluations and advisor networks to identify early-stage targets before they reach the broader market. As a result, Motherbrain supports deal origination as well as ongoing deal management across EQT’s global operations. AI is not only reshaping deal sourcing but also transforming due diligence processes. It can process large volumes of financial statements, legal contracts, and market reports, while also extracting risk factors, identifying inconsistencies, and stress-testing investment theses. PwC further notes that AI can simulate business scenarios prior to an acquisition, making it a particularly powerful tool in pre-deal analysis. In addition, AI can analyse past investment committee materials, anticipate likely questions, and thereby improve deal team preparation. These developments are also influencing the data room provider market, where vendors are increasingly offering AI-powered solutions designed to streamline transaction processes and direct attention to the most material issues. However, AI-driven sourcing and diligence are

Page 2

not without limitations. Data quality remains a fundamental constraint, as AI models are only effective as the inputs they receive. EQT, for instance, embeds data scientists into its due diligence teams specifically to assess the data maturity of target companies. This highlights how even sophisticated adopters must invest heavily in data infrastructure before AI can deliver meaningful value. Beyond availability, integration remains a key challenge. S&P Global found that only 7 percent of PE firms have fully integrated AI technology into their operations, with the majority still in nascent adoption stages. For mid-market firms without the resources to build bespoke platforms, off-the-shelf tools offer efficiency gains but limited differentiation, as competitors access the same technologies and leave relationship networks as the primary edge. Even where AI is successfully deployed, its limitations are evident. Concerns surrounding algorithmic bias, data governance and the opacity of AI-driven recommendations underscore the importance of human judgment. Especially in evaluating strategic fit and relationship dynamics, this judgement cannot be easily quantified.

(3) AI-Driven Portfolio Value Creation As traditional value creation levers such as multiple expansion and financial leverage have become less effective in the current macroeconomic environment, PE firms are increasingly turning to AI as a tool for driving operational improvements within portfolio companies. According to McKinsey’s 2026 Global Private Markets Report, AI is raising the stakes across the entire ownership lifecycle, forcing sponsors to re-underwrite their portfolio strategies with AI-enabled initiatives that can deliver portfolio-wide impact. AI-driven value creation in portfolio companies takes several distinct forms: cost reduction through automation, revenue enhancement through AI-enabled products, and strengthening the exit narrative by demonstrating AI maturity to buyers. While most firms remain focused on the first, the real alpha likely comes from the latter two. A notable example of firm-level AI orchestration is Vista Equity Partners, which has built an “Agentic AI Factory” to deploy AI systematically across its 90-plus enterprise software portfolio companies. Vista reports that more than half of its companies are already monetising AI-enabled products and features, with 30 companies actively generating revenue from agentic AI applications. The firm anticipates deploying between 5 and 10 AI agents per user across its portfolio, translating into billions of autonomous agents handling tasks from customer renewals to code generation. Vista’s portfolio companies are seeing productivity gains of 30 to 50 percent in software development, with some tasks that previously required hours now completed in seconds. Similarly, Bain’s 2025 Global Private Equity Report found that a majority of portfolio companies surveyed across $3.2 trillion in assets under management were in some phase of generative AI testing and development, while nearly 20 percent had already operationalised AI use cases and were delivering concrete results. The key emphasis across leading GPs is that AI must serve strategy, not replace it, and that portfolio companies should focus AI efforts on their most important strategic priorities rather than pursuing unfocused experimentation. Nevertheless, scaling AI within portfolio companies presents significant challenges. A McKinsey survey of PE operating partners found that while approximately 60 percent reported portfolio companies adopting generative AI, only around 5 percent were in production at scale. FTI Consulting reports that 40 percent of PE firms still manage AI investments at the individual portfolio company level under a decentralised model, which limits the ability to share learnings and scale solutions across the portfolio. Change management and employee resistance, often described as “organ rejection,” remain persistent obstacles. Moreover, the transformational benefits of AI require firms to look beyond incremental efficiency gains and consider how AI can fundamentally reshape business models, including how companies sell, what they sell and how they create products and services.

Page 3

(4) AI in Fundraising, Investor Relations and Firm Management Fundraising has become increasingly challenging in recent years, with total capital raised falling 35 percent since 2023, according to PwC. In this competitive environment, AI is emerging as a valuable tool for GPs seeking to identify untapped investors, personalise outreach and improve conversion rates. Leading firms deploy AI models that mine historical fundraising data, LP investment histories and market trends to identify the most promising prospects, such as family offices that have recently gained liquidity from portfolio exits or pension funds seeking exposure to specific strategies and geographies. Beyond prospecting, AI is also streamlining investor relations and fund administration. EQT, for example, has been using an AI-based cash flow forecasting model for liquidity management since 2022, complemented by dynamic KPI dashboards that enable real-time representation of key metrics such as IRR, multiples and EBITDA development. Blackstone has used machine learning models since at least 2021 to simulate fundraising quotas and calculate commitment probabilities, enabling more strategic control of the fundraising process. These tools enable GPs to deliver the precise and timely reporting that LPs increasingly demand, particularly around ESG compliance and performance attribution. The rise of agentic AI, systems that operate proactively and execute multi-step tasks with little human oversight, looks set to push this transformation further. At the firm management level, AI agents are already being assigned specific roles, from sourcing agents that own deal flow identification to research agents that maintain competitive intelligence, effectively mirroring the structure of human teams. However, the integration of AI into fundraising and firm management also raises concerns. An EY survey found that 50 percent of PE respondents believe generative AI and agentic AI will be the most transformative forces in the industry over the next three years, yet the risks of algorithmic bias, lack of transparency and data protection challenges remain material. In investor relations specifically, there is a real question of how far agents can go before they erode the personal contact that LPs value and that has traditionally defined the GP-LP relationship. LPs may also question the reliability of AI-driven forecasts and reporting. As these technologies mature, firms will need robust governance frameworks, clear accountability structures and explainable AI systems to maintain investor confidence and regulatory compliance. PE fundraising remains fundamentally a relationship business. An AI model can identify that a family office recently gained liquidity, but the allocation decision still depends on trust, track record and personal connection. Over-reliance on AI-driven outreach risks making LP communication feel more generic, not less, the opposite of what fundraising requires.

(5) Conclusion Artificial intelligence is no longer experimental in private equity, but its strategic value varies significantly across firms. While AI adoption is widespread, only a small number of firms have scaled it effectively, limiting its impact. This uneven adoption suggests that AI may widen the gap between leading firms and the rest, rather than level the playing field. As core AI capabilities become more widely available, they are increasingly table stakes; competitive advantage will depend on proprietary data and how deeply AI is embedded into investment processes and portfolio strategy.