Greylock's Saam Motamedi: AI creates most disruptive software opportunity since 2005 cloud wave — new pricing, interface, and data models arriving simultaneously

Summary

Greylock partner Saam Motamedi makes a clear-eyed case that the current AI moment represents the most structurally disruptive shift in enterprise software since the 2005 cloud wave — and that the opportunity is being underpriced by incumbents who are protecting near-term margins at the expense of long-term positioning.

The Three-Condition Framework for Software Disruption

Motamedi argues that a genuine software platform shift requires three simultaneous changes: a new pricing model, a new underlying interface, and a new data model. The last time all three aligned was 2005, when cloud computing reset the enterprise software stack. That convergence is happening again now, driven by agentic AI.

The current shift moves pricing toward outcome-based models, interfaces toward agent-driven automation, and data models away from structured schemas toward raw unstructured data. Mobile, by contrast, only changed the interface — which is why it never produced the category-level disruption that many anticipated roughly eight years ago.

Incumbents Are Caught in a Margin Trap

Large public SaaS companies face a structural dilemma. Leaning into AI requires accepting near-term margin compression with the expectation that unit costs will eventually fall and margins recover. Most are unwilling to make that trade, prioritizing short-term profitability over long-term competitiveness. That hesitation is creating clear attack surface for startups.

Motamedi points to Resolve AI, a Greylock portfolio company that automates software operations across observability tools using agents, as an example of the threat vector. When agents replace human end users, incumbent SaaS vendors lose direct relationships with buyers — a structural disintermediation that undermines long-term revenue durability regardless of current retention metrics.

Greylock's Model: Concentrated, Deliberate, SPV-Free

Greylock, which turned 60 years old in 2024 and is now deploying its 17th fund, runs a deliberately simple capital structure. Each fund is a $1 billion vehicle concentrated into approximately 25 core positions. The firm does not use SPVs, viewing them as a distraction from the only activity that actually drives returns — leading first or second rounds in high-conviction companies.

Historical anchors include Palo Alto Networks and Workday, both of which started at adjacent desks in Greylock's San Mateo office roughly 20 years ago. Palo Alto Networks now carries a market cap of approximately $150 billion.

Greylock Edge Targets Enterprise at the Formation Stage

The firm's Greylock Edge program formalizes its company-building capability into a structured offering for pre-seed and seed-stage founders. Nine senior recruiters place an engineer at a portfolio company every other day. A customer development team sources between 40% and 70% of the first two years of pipeline for participating companies. The program is explicitly focused on horizontal enterprise software targeting the roughly 20,000 to 21,000 companies globally that generate more than $1 billion in annual revenue — the cohort that controls the majority of global IT spend.

VC Market Structure Is Reconcentrating

The explosion of solo GPs and new funds during the 2021 period is now reversing. Motamedi expects the total number of active venture firms to decline as power law dynamics intensify under AI — fewer winning companies, fewer firms with access to those companies, and larger fund sizes at the top. Corporate capital, particularly from Nvidia, is absorbing a meaningful share of late-stage AI financing that would not historically have flowed through traditional venture structures. Crossover funds, debt facilities, and strategic capital are filling similar roles, partially explaining why aggregate venture fundraising data looks depressed even as headline deal sizes grow.