TMT Conference Day 3: AI Agents Shift to Enterprise Deployment, Enabling Infrastructure Value Underappreciated

Core Conclusion AI agents are transitioning from conceptual exploration to tangible enterprise deployment, with monetization expected to accelerate in 2026. The critical insight from industry leaders is that sustainable value creation lies not in the foundational models themselves, but in the enabling infrastructure—orchestration, security, governance, and high-quality, structured data context—required to make agents reliable for mission-critical workflows. Companies with deep enterprise embeddings, proprietary data, and deterministic system control are building durable moats and pioneering new pricing models to capture this value.

Market Mispricing The market likely misprices the locus of value in the AI agent stack, overweighting model providers while underweighting the enabling infrastructure layer. Enterprise adoption hinges on orchestration, security, governance, and integration with high-quality, often proprietary, data. These components act as the control plane and bottleneck for reliable agent deployment, creating economic value through higher pricing, longer contract terms, and stronger platform lock-in—a value currently underappreciated.

Evidence Chain 1. Agent deployment is accelerating, driving measurable pricing power and platform transitions. Concrete business outcomes are emerging. ServiceNow's Now Assist has scaled to approximately 3,000 customers, sustaining pricing uplifts around 30%, with reload activity up 55x since May 2025. BlackLine's transition from seat-based to platform pricing is exceeding expectations; eligible ARR on the new model grew from ~4% to ~11% in a single quarter, targeting 25-35% by end-2026 with an average uplift exceeding 10%. The investment implication is a re-rating of the infrastructure software and workflow platforms enabling this shift, as early monetization signals validate the durability of their role.

2. The enabling infrastructure layer is emphasized as the core source of differentiation and value. Executives consistently highlighted that agents require a robust orchestration and control foundation. Microsoft's strategic moat is its M365 data layer, WorkIQ, which acts as a stateful context engine enabling agents to reason across workflows—a key differentiator for CIOs focused on security and compliance. Dynatrace's CEO argued enterprises will not trust "probabilistic input → probabilistic output" for critical workloads, necessitating its platform's deterministic observability as the essential control plane. This reinforces that competitive barriers are built on integrated data context and reliable orchestration, not just agent capabilities.

3. Pricing models are actively evolving towards hybrid structures to capture agent economics. Companies are moving beyond pure seat-based licensing to align pricing with the value delivered by automation. Microsoft's CEO expects business models to blend subscriptions for predictability with usage-based metering. Freshworks is intentionally evolving its monetization to a mix of seat-based copilots, consumption-based agents, asset-based pricing, and incident-based pricing. This pricing innovation is a critical precursor to the anticipated 2026 monetization acceleration, expanding TAM and ARPU potential for successful adopters.

Key Divergence & Risks

• Execution & Adoption Risk: Integrating complex agent orchestration layers and messy enterprise data is operationally challenging, potentially delaying deployments or limiting realized value.
• Pricing Transition Friction: Shifting from traditional seat-based models to agent/consumption-based pricing may face customer resistance, sales compensation complexities, and revenue recognition headwinds in the short term.
• Macro Uncertainty: A deteriorating economic environment could slow budget allocation decisions for new agent functionalities, delaying the 2026 monetization timeline.

Valuation & Trade Implications Valuations should increasingly reflect advantages in: 1) Infrastructure & Orchestration (providing the control plane for agents), 2) Proprietary Data & Workflow Context (owning deterministic, mission-critical systems), and 3) Successful Pricing Transition (moving to hybrid/consumption models). Investors should prioritize tracking concrete metrics like pricing uplift percentages, platform ARR mix, and contract duration extensions over generic AI feature announcements. Companies demonstrating an ability to translate the agent trend into stronger pricing power and customer lock-in warrant premium multiples.

Appendix: Select Agent Monetization & Adoption Metrics