Companies buying artificial intelligence services are finding it difficult to keep spending under control, while the firms selling those tools are still working out what a fair price looks like. The result is a market where both sides want clarity, but neither side has much certainty.
A key problem is that AI pricing is still changing too quickly for traditional long-term contracts. Simon Gooch at Saviynt said trying to lock someone into a cost model for the next 12 months, two years or three years does not really make sense because the industry still does not know enough. That reflects how unsettled the economics of AI remain.
For customers, this creates budgeting problems. Businesses want predictable costs before committing to new software or services, especially when AI can be used across many teams and projects. But if pricing structures are still shifting, it becomes harder to compare offers or plan spending over time.
For sellers, the challenge is different but just as significant. AI providers need to decide how to charge for tools whose usage, value and delivery can change rapidly. The report underlines a basic tension in the AI market: buyers want control over costs, while vendors are still figuring out how to price a fast-moving technology.
