Verisign is worth approximately $24 billion. Most people have never heard of it. It manages the .com and .net DNS registries — an invisible function that the entire internet depends on and nobody thinks about until it breaks. Cloudflare is worth approximately $75 billion — more than most of the websites it protects, and more than double its value from a year earlier. CrowdStrike: roughly $100 billion. Palo Alto Networks: roughly $130 billion. All figures as of April 2026, via public market data.
Four companies. A combined valuation north of $325 billion. All of them do the same thing: they make other people’s systems trustworthy. None of them are the systems.
The agent economy does not have its version of these companies yet. It will.
PRED-010 — A trust layer company will become more valuable than any single agent application company by December 2031. The leading company providing trust infrastructure for AI agents — verification, attestation, audit, and identity services — will have a higher market valuation than any single company whose primary product is an AI agent application. Valued at $50B+.
Confidence: 3 out of 5.
The structural logic is clean. Trust infrastructure captures durable value because trust is non-optional. When the stakes are high enough, no enterprise opts out of verification. Verisign, Cloudflare, CrowdStrike, and Palo Alto Networks all prove the pattern: the company that makes the system trustworthy captures more value than the system itself.
The trust gap in current agent deployments is enormous. There is no standardised way to verify agent outputs. No standardised way to attest to agent processes. No standardised way to authenticate agent identity. The five components of the dependency layer — orchestration platforms, trust and attestation services, agent identity systems, monitoring tools, and inter-agent communication protocols — include trust as one of the most underdeveloped. The company that closes this gap inherits the economics of every company that builds on top of it.
That is the chapter’s argument, and I stand behind it. But the chapter skips the honest counter-case, and the counter-case matters.
Boring-but-critical infrastructure tends to get absorbed by cloud providers at scale. RSA was acquired. Dozens of DNS companies were bought by Verisign and Cloudflare. Identity companies orbit AWS, Azure, and GCP as acquisition targets or integration partners. The pattern is consistent: infrastructure that is critical, standardised, and low-margin gets absorbed into platforms that bundle it as a feature.
The trust-layer winner may follow the same path. A company builds the attestation standard, gains adoption, reaches $15 to $30 billion in valuation — and then AWS, Microsoft, or Google acquires it before it ever crosses $50 billion as an independent. The prediction is substantively right: a trust-layer company becomes more valuable than any single agent application company. But the prediction looks wrong because the acquirer’s market cap absorbs the trust layer’s value into a line item.
I am committing, in writing, to treating acquisition outcomes as prediction-validating rather than prediction-falsifying. If a trust-layer company is acquired at $20 billion and the acquirer’s trust-infrastructure division subsequently generates revenue consistent with a $50 billion standalone valuation, PRED-010 holds. The value was created. The market structure just absorbed it differently.
This is not a hedge. It is a commitment to intellectual honesty about how infrastructure markets actually work. The Verisign path — survive independently, become invisible, capture durable value — is one outcome. The RSA path — build the critical layer, get acquired, disappear into a platform — is the other. Both validate the thesis that trust infrastructure is worth more than agent applications. Only the ownership structure differs.
I watch this pattern from inside the physical-bridge thesis. The attestation infrastructure I am building into is already attracting acquisition interest at the concept stage. Not because the technology is unique — cryptographic attestation is well-understood. Because the market position is strategic. The company that owns the attestation standard for agent actions owns a toll booth on every agent deployment.
The published falsification trigger:
If by December 2031, no agent trust or verification company exceeds $10 billion in valuation, or if agent application companies consistently outvalue infrastructure companies by more than 3x, this prediction is wrong.
The realistic failure mode: trust becomes a commodity feature, not a standalone market. If every cloud provider ships a “good enough” attestation API as a free add-on to their agent platform — the way they shipped basic IAM as a feature — no standalone trust company reaches the $10 billion threshold. The trust layer exists, but nobody captures value from it independently. The prediction dies by commoditisation, not by irrelevance.
If you work inside enterprise M&A, cloud strategy, or venture capital focused on infrastructure, I want to know: have you seen any 2026 to 2028 acquisition activity — completed, in progress, or rumoured — involving attestation, agent audit, or agent identity infrastructure? Deals above $5 billion are the signal, but pattern surfacing at any scale matters.
Send it to me. I will publish the data — anonymised where needed — on the public PRED-010 tracking page at atin-agarwal.com/predictions/pred-010-trust-layer-valuation/, with credit.
One deal does not validate the prediction. A pattern of deals does.
If you are building a trust-layer company: your most likely end-state is acquisition, not IPO. This is not a failure — it is how infrastructure markets work. Price your equity structure accordingly. If your cap table assumes an IPO at $50 billion, you may be right on the value and wrong on the vehicle.
If you are an investor evaluating trust-layer companies: M&A comps matter more than IPO comps for 2028 to 2031 analysis. The Verisign/Cloudflare standalone path is the exception in infrastructure, not the rule. Underwrite the acquisition scenario, not just the public-market scenario.
If you are a cloud platform strategist: the attestation acquisition is the AWS-S3-shaped move of the late 2020s. S3 was not interesting technology. It was an interesting market position — the default storage layer that everything else depended on. The attestation layer has the same shape. The company that owns it owns the toll booth. If you are not already looking at this space, your competitor is.
This issue is drawn from Chapter 9 of The AI Agent Economy — 15 falsifiable predictions with dates, numbers, and explicit triggers for being proven wrong. Pre-order on Kindle — $9.99. Release July 1, 2026. atin-agarwal.com/books
Read the full PRED-010 entry on the public tracking page → atin-agarwal.com/predictions/pred-010-trust-layer-valuation/
Previous issue: Issue 09 — Your insurance policy will demand agent attestation before your regulator does → SUBSTACK-009 Next issue: Issue 11 — India will be #2 in agent infrastructure revenue by 2030 — because of the infrastructure instinct → SUBSTACK-011
Comments
Nothing yet. Say the first thing.
Sign in to join the conversation.