Artificial intelligence is changing how private equity and venture capital firms assess investments and monitor portfolios, but human judgment, proprietary relationships, and data remain critical to investment decisions, investors said at the Asia PE-VC Summit in Singapore.
AI is increasingly being used to speed up due diligence, analyse unstructured data and challenge investment theses, said Yi Pan, principal at Neuberger.
A survey of more than 140 general partners conducted by Neuberger Private Markets found about 90% were increasing AI spending in due diligence, while about 78% were increasing spending on AI in the investment decision process, he said.
Pan added AI could help investment committees build conviction by providing additional perspectives and broader analysis, but it remained a decision-support tool rather than a replacement for human judgment.
AI has been used to analyse data that would traditionally be difficult to process, including thousands of employee feedback notes when assessing the potential merger of two businesses, said Charles Allen, senior adviser for digital operations at TPG Asia.
The analysis helped investors identify cultural differences, employee expectations and other factors, allowing the deal team to develop a more specific integration plan, he said.
“AI not only speeds us up, but also allows us to provide new perspectives on deals,” Allen said.
Investors said the quality and availability of data remain important constraints, particularly in Southeast Asia, where information on some private companies can be limited.
Xuan Ye, managing director at multi-family office Collyer Capital, said AI adoption among private equity firms varies depending on their resources. Larger funds can approach AI more systematically, while smaller managers may use it more on an individual basis.
But AI cannot replace relationships with founders and family-owned businesses, particularly in Southeast Asia, where deal sourcing often relies on industry knowledge and personal networks, Ye said.
“AI is not going to bring in good deals,” Ye said, adding that sourcing remains based on “human relationships” and the ability to connect with founders and family businesses.
Pan said AI could help LPs analyse historical data when assessing fund managers, but factors such as team quality, incentives, retention and the ability to execute a strategy still require human assessment.
“Human judgment is actually core to underwriting GPs in this process,” Pan said.
Basil Lui, founding partner and CEO of August Global Partners, said private equity firms retain an advantage through proprietary information gathered through board meetings, relationships, and other interactions that are not captured in conventional datasets.
“Where PE/VC shines is the private proprietary data that the AI can never scrape,” Lui said.
Lui said his firm uses an AI system to screen about 2,000 potential companies each night, narrowing the pool for further due diligence and allowing the investment team to focus on a smaller number of companies that meet its criteria.
He said AI could potentially perform work that would previously have required a much larger analyst team, while investment decisions would remain with senior professionals.
At TPG, AI is increasingly being incorporated into underwriting as one of the potential levers for improving a company’s growth trajectory, Allen said.
The firm has also developed an internal AI tool that draws on its historical investment experience to provide deal teams with a retrospective perspective on new investments, although it is not itself part of the decision-making process.
Pan said AI governance would be as important as AI capability as firms integrate the technology into investment decisions, with accountability ultimately remaining with human deal teams.
Allen said TPG expects its traditional staffing structure to remain in place, while employees’ roles evolve as AI takes on more analytical work.
“We’re now seeing expectations around what each level does and how it gets AI augmented,” he said.
The panel also highlighted the risk of investors overestimating AI’s effect on earnings.
Ye said he had seen companies with negative returns on AI investments overstate the technology’s potential to attract customers and increase revenue.