Key Takeaways

  • David Sambur, co-head of private equity at Apollo, structures AI deployment across three clear tracks: portfolio operations, investment evaluation, and internal firm productivity.
  • Apollo gathers portfolio leadership twice a year for dedicated symposiums, treating its collection of companies as a shared test kitchen to test and scale software tools.
  • Deal teams now run custom prompts and proprietary agents directly inside underwriting workflows to evaluate potential acquisitions from fresh angles.
  • Firm operations have shifted from retrospective quarterly reporting decks to live, screen-based performance tracking.
  • Apollo organizes these operational interventions through Apollo's Three-Pillar AI Deployment Framework.

Apollo's Three-Pillar AI Deployment Framework

Sambur outlines how Apollo divides its technology investments to capture upside while stripping out administrative friction across its portfolio:

  • Pillar 1: Portfolio Company Operations: Deploying specialized teams and use-case playbooks across portfolio businesses in targeted areas like automated customer service, digital marketing, real-time operations, and pricing, using portfolio companies as a collaborative test kitchen.
  • Pillar 2: How We Invest: Integrating custom agents, structured prompts, and automated analytical tools directly into deal underwriting, diligence workflows, and investment decision-making.
  • Pillar 3: How We Work Internally: Applying AI as a productivity force multiplier to automate routine corporate workflows, administrative compliance, quarterly reporting drudgery, and live data dashboard tracking.

When This Works (and When It Doesn't)

This framework works best for mega-cap sponsors managing massive, cross-sector asset bases where operational experiments create compounding data network effects. Apollo runs semi-annual gatherings with its operating leaders to share what works in content creation, pricing, and customer support. When one business finds an automated edge, the sponsor transfers that playbook across the rest of the portfolio within weeks. As Sambur explains: “The portfolio companies are a great test kitchen for us to try stuff.”

Where this model breaks down is inside mid-market firms with lean operating teams and heterogenous tech stacks. Without standardized data architectures or centralized operating teams, portfolio companies rarely share insights across silos. The model also fails if investment teams treat custom prompts as automated rubber stamps for diligence. Sambur stresses that AI is “not a substitute. It's a force multiplier.” It sharpens human judgment during underwriting, but it cannot replace sponsor conviction or deep domain diligence on messy targets.

Why It Matters

Apollo's operational playbook signals a clear divergence in private equity value creation strategies. Sponsors can no longer rely on multiple expansion or low borrowing costs to hit baseline return thresholds. Instead, returns increasingly depend on immediate operational interventions that boost EBITDA from day one. By treating portfolio holdings as a linked testing ground, large sponsors build proprietary execution advantages that single-asset owners cannot replicate. At the firm level, shifting from backwards-looking quarterly reports to live performance dashboards compresses the feedback loop between operational drift and sponsor intervention, giving institutional managers faster visibility into capital allocation decisions.