Accenture cut its guidance and cited "AI demand compression" — a phrase now defining a sector-wide crisis across US, European, and Asian enterprise software markets.1 The synchronized decline across vendors signals a global ROI credibility problem, not isolated company failures.
Enterprise procurement teams from Frankfurt to Singapore to Chicago are asking the same question: what did two years of AI tooling spend actually return? For most organizations, the answer remains unclear.
Accenture's use of "AI demand compression" in formal earnings language — not a passing remark — suggests leadership views the slowdown as structural. Companies do not introduce new vocabulary into guidance without deliberate intent.
The pressure is spreading into services. IBM, Infosys, and Cognizant — the last two India-headquartered IT majors with global delivery footprints — face equivalent exposure as enterprise clients worldwide hold discretionary AI budgets.1 Software vendor softness historically precedes parallel contraction in services revenue.
One to two quarters of further multiple compression are likely for this segment.1 Enterprise buyers are not abandoning AI — they are demanding proof of returns before extending commitments.
A clear divide is emerging between two categories of AI spend. Infrastructure investment — compute, data pipelines, model access — continues expanding globally. Application-layer and consulting spend, where ROI is harder to quantify, is stalling across markets.
For Salesforce and Adobe, the challenge is universal: convert AI feature adoption into measurable business outcomes. Adobe's generative tooling and Salesforce's agentic platform face identical pressure whether the client sits in London, Tokyo, or New York.
Atlassian's smaller decline relative to peers may reflect its tighter integration into developer workflows, where productivity gains are easier to instrument — a lesson for vendors across geographies seeking budget approval.
The next one to two quarters will determine whether this is a temporary pause or a permanent repricing of enterprise AI. Vendors who can surface concrete ROI data — cost savings, cycle time reduction, measurable output gains — will emerge at a lower discount than those who cannot.
