New survey finds 72% of leaders are confident in AI outputs, yet 65% still manually validate or rework them, and fewer than 3 in 10 can always explain the data behind an answer
Most enterprises have paid for AI once. Now they are paying for it again. Stibo Systems, a global leader in master data management (MDM), today released The Trust Tax, a new research report showing that enterprises are paying a hidden second bill for AI: the validation, reconciliation and rework needed before anyone will act on what AI produces.
AI adoption has outrun trust. Four in five leaders (80%) have scaled AI across functions or embedded it in core business processes, but the data behind it is not yet reliable enough to act on without a human double-checking the work. Stibo Systems calls this the Trust Tax.
Key findings
- The trust paradox: 72% of respondents report high confidence in AI outputs, yet 65% often or always manually validate, check or rework them.
- Hidden labor cost: Four in five spend at least six hours a week preparing, reconciling or validating data for AI and analytics.
- Explainability gap: Only 29% say they can always explain the data behind an AI output.
- Confidence is built on data: High-confidence organizations are far more likely than moderately confident peers to have unified data (93% vs. 70%), traceable data (41% vs. 10%), a centralized AI governance council (25% vs. 16%) and a true MDM or enterprise data foundation (53% vs. 42%).
- Trust pays off: High-maturity organizations are more than twice as likely to report strong positive ROI from AI (31% vs. 14%).
- Why companies invest: Operational efficiency (51%), employee productivity (40%), customer experience (35%) and faster decision-making (33%) lead the motivations for AI spend.
What it means
The research points to a clear fix: human oversight is not the problem; disorganized data is. Organizations should keep people in the loop on high-stakes decisions while eliminating the rework caused by incomplete or unexplainable data. And as companies move toward AI agents, the lesson holds: agents need better evidence.
“Enterprises have invested heavily in building AI, and now they’re paying a second bill in validation, reconciliation and rework,” said Gustavo Amorim, Stibo Systems Chief Marketing Officer. “AI creates intelligence, but confidence is what executives require, and verification is what their teams are doing. The answer is trustworthy intelligence: data that is clean, governed, contextualized, and explainable. Without it, there is no real return on AI investment.”
The report shows that confidence in AI tracks data governance maturity more than model sophistication. Organizations that can explain, trace and trust the data behind every output are the ones reporting the strongest returns.
The Trust Tax findings are available here.
The post Stibo Systems Research Reveals the Hidden AI Trust Tax on Enterprises first appeared on AI-Tech Park.
