‘Excited and terrified’: One of private equity’s top investors built an AI that knows every deal he’s ever done

Advent International is actively deploying AI across investing and portfolio operations, centered on an internal 'IC Robot' trained on 13 years of investment committee materials to challenge new deal assumptions. The firm has also built its own LLM stack, hired data leadership early, and is embedding AI staff in portfolio companies to drive transformation. The article is strategic and informative rather than event-driven, so immediate market impact appears limited.

Analysis

The key signal is not that AI is improving diligence, but that top-tier private equity is turning prior deal history into a machine-readable feedback loop. That should widen the gap between scaled platforms with proprietary data and everyone else: better-funded buyers will underwrite faster, reject more weak deals, and selectively pay up for situations where AI can extract operating leverage. The second-order effect is more pressure on smaller sponsors and bankers who rely on pattern recognition and relationship edge rather than data infrastructure; their hit rate should deteriorate first in software, industrial carve-outs, and complex cross-border deals.

For Goldman, the near-term implication is less about direct AI monetization and more about advisory and financing share. If large sponsors professionalize diligence and portfolio transformation with in-house AI, they should become more aggressive in sourcing, more confident in process control, and more willing to run broader auctions with tighter decision timelines. That tends to favor the few banks that can offer integrated M&A, financing, and data-heavy execution, while commoditizing mid-tier advisory and making “relationship-only” pitches less defensible over the next 12-24 months.

The contrarian risk is that this is still mostly a governance and workflow story, not an immediate earnings inflection. Private equity returns are constrained by implementation, culture, and data cleanliness; AI will likely raise the floor on mediocre decisions before it materially lifts the ceiling on great ones. The market may already be overpricing a near-term productivity miracle, but underpricing how sticky the operating-model change becomes once firms reorganize around structured data and AI-native hiring. If that happens, the winners will be the platforms that can convert better diligence into faster deployment and lower loss ratios, not merely the ones with the flashiest AI narrative.

Key Decisions for Investors

• Long GS vs. short KKR/HLI on a 3-6 month horizon: GS has cleaner leverage to AI-enabled advisory and capital-markets share if large sponsors become more execution-heavy; the short leg captures the risk that AI commoditizes lower-tier advisory economics faster than the market expects.
• Buy GS call spreads 6-9 months out, financed with shorter-dated premium if vol stays elevated: positioning for a modest re-rating in investment-banking activity and tech-enabled client engagement, with limited downside if AI remains mostly narrative in the near term.
• Long XLK / short XLB over 6-12 months: AI-driven diligence and portfolio optimization should favor software/data winners more than asset-heavy cyclicals that face margin pressure from better capital allocation and faster competitor learning.
• Watch for a basket long in defense-tech/industrial automation enablers on any pullback: firms with data-rich operating environments should be the first beneficiaries of AI deployment inside PE-owned assets, but entry should wait for confirmation that portfolio spending is translating into productivity gains.
• Avoid chasing pure-play AI consultancy names here; if the thesis is correct, the value accrues to owners of proprietary workflows and data, not generic implementation services. Use rallies to fade firms without embedded distribution or differentiated data moats.