Key Takeaways
- • Veritas Capital Managing Director Nicolas Schaufele said access to AI models is now universal across private equity, calling it "beta" and arguing that alpha comes from proprietary portfolio-company data and the ability to harvest it.
- • Vista Equity Partners SVP Hannah Dolgin said Vista's edge lies in "access to context, trust, and scale," citing portfolio company Duck Creek, which supports underwriting for property and casualty insurers.
- • Schaufele cited a Veritas manufacturing portfolio company where AI cut the new-product launch timeline "from two years to six months" and raised yields "by up to 30%."
- • Panelists agreed revenue gains from AI are less proven than cost savings, with Schaufele noting it is unclear "how much of it is AI driven versus just other innovations."
- • Tikehau Capital's Cécile Mayer-Lévi said AI is now part of credit underwriting at Tikehau, with every deal assessed for whether AI is a risk or an opportunity.
Summary
Executives from Vista Equity Partners, Veritas Capital and Tikehau Capital argued that proprietary data, workflow context and human judgment, rather than access to AI models, will determine which private equity firms outperform, according to a panel at the Israel Private Markets Summit (IPM) held on Tuesday at the Peres Center for Peace and Innovation in Tel Aviv, as reported by CTech. Moderator Ofir Reiner Sigler, Head of Direct and PE Investments at Menora Mivtachim Group, framed the discussion as a split between "AI alpha" and "AI beta."
Nicolas Schaufele, Managing Director at Veritas Capital, said, "Everybody has access today. I think that's beta," adding that integration "will commoditize over time." He argued that alpha comes from proprietary data inside portfolio companies and from knowing how to harvest it. Hannah Dolgin, Senior Vice President at Vista Equity Partners, made a similar case, saying "what really sets us apart is the access to context, trust, and scale." She cited Duck Creek, a Vista portfolio company supporting underwriting for property and casualty insurers, noting that it is "not giving that information away to general purpose models."
On surprises, Schaufele said hallucinations, which he would have named a year ago, are "mostly solved," and that the remaining gap lies with firms, including workflows, willingness to adopt AI and how data is structured. Dolgin said the speed of adoption has been "much faster than we've historically seen" over Vista's 25-year history. Schaufele described AI-driven retooling at a Veritas manufacturing portfolio company that cut product launch time "from two years to six months" and increased yields "by up to 30%."
Cécile Mayer-Lévi, Head of Private Debt at Tikehau Capital, said AI is now part of underwriting at the firm. On cost cutting, she cautioned that savings are not automatic, citing a large insurer that deployed a copilot tool, saw no efficiency gains and switched it off. Dolgin said cost savings are a given, with all Vista portfolio companies running AI cost-saving initiatives, while revenue growth is less proven. Vista's strategy, she said, includes tapping labor spend as an addressable market and using its "agentic factory" to place engineers inside portfolio companies. Schaufele said revenue is accelerating where companies hold data assets but remains unclear in attribution.
On junior professionals, Mayer-Lévi said the era of lengthy presentation work is ending, while Schaufele expected juniors to be exposed to judgment earlier. Dolgin noted that models produce probabilistic outcomes, making human oversight essential, and Schaufele said firms should treat models like a red team and review past mistakes together. "You need to be careful and not just rely on the models blindly," he said.
Nicolas Schaufele, Managing Director at Veritas Capital, said, "Everybody has access today. I think that's beta," adding that integration "will commoditize over time." He argued that alpha comes from proprietary data inside portfolio companies and from knowing how to harvest it. Hannah Dolgin, Senior Vice President at Vista Equity Partners, made a similar case, saying "what really sets us apart is the access to context, trust, and scale." She cited Duck Creek, a Vista portfolio company supporting underwriting for property and casualty insurers, noting that it is "not giving that information away to general purpose models."
On surprises, Schaufele said hallucinations, which he would have named a year ago, are "mostly solved," and that the remaining gap lies with firms, including workflows, willingness to adopt AI and how data is structured. Dolgin said the speed of adoption has been "much faster than we've historically seen" over Vista's 25-year history. Schaufele described AI-driven retooling at a Veritas manufacturing portfolio company that cut product launch time "from two years to six months" and increased yields "by up to 30%."
Cécile Mayer-Lévi, Head of Private Debt at Tikehau Capital, said AI is now part of underwriting at the firm. On cost cutting, she cautioned that savings are not automatic, citing a large insurer that deployed a copilot tool, saw no efficiency gains and switched it off. Dolgin said cost savings are a given, with all Vista portfolio companies running AI cost-saving initiatives, while revenue growth is less proven. Vista's strategy, she said, includes tapping labor spend as an addressable market and using its "agentic factory" to place engineers inside portfolio companies. Schaufele said revenue is accelerating where companies hold data assets but remains unclear in attribution.
On junior professionals, Mayer-Lévi said the era of lengthy presentation work is ending, while Schaufele expected juniors to be exposed to judgment earlier. Dolgin noted that models produce probabilistic outcomes, making human oversight essential, and Schaufele said firms should treat models like a red team and review past mistakes together. "You need to be careful and not just rely on the models blindly," he said.
Go deeper than the headline
Research every firm in this story, track the ones that matter and get the daily briefing — free for 7 days.