Enterprise software companies built their businesses around one assumption that went unquestioned for decades. People are the users. Employees log in, click through interfaces, update records and move between applications. The software earns its seat license because a human being sits in front of it.
That assumption is starting to crack. AI agents are becoming capable of retrieving information, making decisions and executing workflows on behalf of employees. The question circulating across enterprise software boardrooms is what happens to the business model when the human in the browser tab is no longer the primary user.
Why AI agents could make enterprise platforms more important
Salesforce is already confronting this. In April, the company launched Headless 360 at its TrailblazerDX conference. The architecture exposes every Salesforce capability as an API, MCP tool, or CLI command so that AI agents can access data, workflows, and business logic without opening a browser.
Co-founder Parker Harris framed it directly: “Why should you ever log into Salesforce again?”
That question points at something counterintuitive. The screens may matter less. The systems underneath may matter more.
An enterprise CRM or IT service management platform holds years of accumulated business rules, permissions, workflows, and records. That is exactly what an AI agent needs before it can act. The interface depreciates. The infrastructure underneath it appreciates.
More AI:
- Nvidia just made a move Wall Street wasn’t ready for
- Microsoft just took sides in AI policy fight
- OpenAI just disclosed something genuinely alarming
“The platforms and systems of record are not losing value here,” Yoav Kolodner, CEO of Tribal and a former VP of engineering at Salesforce, told TheStreet.
“If anything, they matter more, because what an agent actually needs is the object model, the business logic that took fifteen years of edge cases to accumulate, the permission model, the audit trail. None of that is easy to rebuild, and all of it is exactly what an agent consumes.”
Tribal is an enterprise AI platform that maps the metadata and dependencies inside systems of record. It lets AI agents operate within existing permissions and governance frameworks.
The readiness problem Kolodner describes is real. An agent that sees three fields called Status, Status_c, and Stage_Old_c will act on whichever one it finds first.
A human admin knows which one is live, but an agent does not.
How AI agents are disrupting enterprise software pricing
The more immediate problem for public enterprise software companies is what happens to revenue. The industry has run on seat-based pricing for years. More employees using a platform means more licenses.
But if AI agents take over tasks that previously required several employees, the relationship between usage and license count gets harder to justify.
ServiceNow has been shifting toward non-seat pricing arrangements, with more than half of its new business coming from consumption and outcome-based deals. That direction is right. The hard part is defining what an outcome actually is.
Much of what vendors currently call consumption pricing is closer to activity billing. Customers are charged per action or per AI credit. That measures how much compute the vendor spent, not how much value the customer received.
A badly built agent that retries five times shows up on the invoice, and the customer pays for the inefficiency.
The vendors who come through this transition will be the ones who tie price to something the customer already tracks. Tickets resolved. Cases closed. Days off a billing cycle, according to Salesforce Ben.
Kan/Getty Images
Where the next enterprise software layer is being built
The shift is creating room for companies that sit between AI agents and the existing software stack. Most large companies will not replace their core systems because a new AI capability becomes available. The switching cost is too high. The customization runs too deep.
That creates a market for infrastructure that helps agents understand and operate across existing enterprise environments. The company builds its platform around a customer-specific map of what data exists, what actions are available, and who is allowed to do what.
Almost no real business process lives in a single application. A sales transaction spans CRM and ERP. An IT incident touches service management, customer records, and HR systems.
AI agents can read across those environments. Acting across them is harder. Permissions and governance rarely align at the seams between platforms.
Why enterprise disruption starts at small businesses first
For investors in established enterprise software, the key question is how fast this transition moves. The AI agent discourse suggests a near-term overhaul. The actual timeline is more complicated.
“My honest view is that in the long run the underlying platforms do get disrupted. If agents become the primary way work gets done, a lot of what a system of record charges for today starts looking like a database with very expensive UI on top,” Kolodner added.
“But that’s a much slower story than the current discourse suggests. Large enterprises don’t switch systems of record on a capability argument. They switch on trust, and trust in this category is going to take years to earn.”
The disruption starts where switching costs are lowest. Small and mid-size businesses have less customization and fewer compliance requirements. Some are already running operations on a thin agent layer with minimal legacy infrastructure.
Enterprise adoption follows once that model is proven. Kolodner puts the full picture five to 10 years out.
For enterprise software investors, the near-term risk is pricing pressure as vendors renegotiate seat-based contracts. The longer-term question is whether companies that built their moats around powerful interfaces can hold that moat when the interface is no longer the point.
