New York and Brussels – September 19, 2026 -- Nearly nine in ten enterprise decision-makers (87%) say their teams routinely burn hours re-verifying that context fed to autonomous AI agents remains accurate, according to a Harris Poll survey of 306 U.S. data, privacy, and AI decision-makers commissioned by Collibra. The report, titled The 2026 Hallucination Tax Report, found that 76% of organizations have hit critical roadblocks trying to move AI agents from pilot programs into full production over the past 12 months.
Poor data foundations drive project failures, executives say
Some 72% of respondents agree that when AI initiatives fall short, the root cause traces back to an unaligned or poor data foundation. Among organizations with $100 million or more in annual revenue, that figure jumps to 96%. Gartner research cited in the report separately estimates that at least 50% of enterprise generative AI projects are abandoned after proof of concept due to poor data quality, escalating costs, and inadequate risk controls.
Manual review drains staff time at large enterprises
Just over half of respondents (51%) report spending significant staff hours manually reviewing and correcting autonomous AI agent outputs before deployment. That burden is more acute at larger firms, where 64% of decision-makers at organizations with $100 million-plus in revenue report manual review drains on staff time.
Companies restructure reporting lines to merge AI and data governance
More than half of decision-makers (53%) say the reporting line for their AI function has moved closer to the primary data organization over the past year, rising to 62% among $100 million-plus organizations. Separately, 84% of organizations report having clearly defined executive accountability for cases when autonomous agents produce flawed or harmful outputs.
Nine in ten firms prepare for tightening AI regulation
Ninety percent of leaders report actively preparing for evolving AI regulations across federal, state, and international jurisdictions. Organizations are prioritizing clear internal accountability for AI outputs and decisions (58%) and targeted investments in data lineage and documentation (51%) to ensure long-term compliance.
"The AI conversation across the C-suite has fundamentally shifted from the theoretical to measurable business impact," said Felix Van de Maele, Co-founder and CEO of Collibra. He said every enterprise scaling AI today pays a hidden "hallucination tax" of manual oversight, rework, and risk that grows with each new agent put into production, adding that the business value of automation disappears if human workers must manually validate every response.
The survey was conducted online by The Harris Poll on behalf of Collibra from August 5 to 11, 2026, among 306 U.S. adults aged 21+ employed full-time as data management, privacy, or AI decision-makers at the director level or higher. The full sample is accurate to within +/- 6.4 percentage points at a 95% confidence level.