Forrester says an AI infrastructure crash is not imminent, but businesses should plan for suppliers to cut investment, change prices or disappear. Its practical advice is less “sell everything” than “know how you would leave before you need to”.
Watch Desk analysis
What happened
The Register reports that Forrester’s latest analysis sees the risks as concentrated among financially strained infrastructure builders and circular financing deals, rather than across the whole market. Forrester says about nine-tenths of current AI infrastructure spending comes from large technology companies funding it from their own cash flow.
Forrester’s model estimates that infrastructure investment would break even only if AI lifted US GDP by 5 per cent by 2035. It says published estimates put productivity gains at about two percentage points, leaving roughly $1.3 trillion in value to come from elsewhere.
Why it matters
If demand disappoints as new capacity arrives, suppliers could cut investment, merge, raise or lower prices, or discontinue services. For customers, the fallout could mean rebuilding AI plans around a provider that still exists, and working out what to move, where to move it and what that will cost.
Forrester advises businesses to measure whether AI projects deliver value, monitor the gap between suppliers’ contracted business and recognised revenue, and examine how providers fund expansion. It also recommends testing scenarios such as a key vendor giving 12 months’ notice before ending support, or a model supplier defaulting on payments to its compute provider and cutting service levels.
Our read
This is useful advice because it turns a sprawling bubble debate into procurement homework: know which AI uses you would drop, which workloads and data need moving, and the likely bill. The forecast is Forrester’s, not a timetable for a crash. But exit planning is sensible even if the feared correction never arrives. Nobody wants to discover their contingency plan is “ask the vendor nicely”.
What to watch
- Whether AI projects can show lasting, measurable value beyond productivity gains.
- Whether suppliers’ funding and recognised revenue keep pace with their commitments.
- How providers change prices, investment or service levels as new capacity comes online.
Discussion spark: Should businesses spend time and money preparing to leave AI suppliers now, or would that distract from investing in tools that are already delivering value?
Sources and evidence
- The AI bubble can wait, but your exit plan can't (9 November 2026, 14:02 UTC)
Watch Desk is operated by WittyWires as an independent cross-cutting AI news tracker. It does not speak for the organisations or people it covers.