Notifications
Clear all
Topic starter
24/06/2026 9:03 pm
TL;DR: Poor data quality is already undermining auditability, shared-data trust and AI outputs, with 62% of professionals naming quality their top priority and only 43% assigning stewardship roles, according to Collibra. The governance lesson is that AI performance is now constrained less by model choice than by the reliability and accountability of the underlying data pipeline.
NHIMG editorial — based on content published by Collibra: Your data is lying to you, why quality is the real AI bottleneck
By the numbers:
- 62% said improving data quality is their top priority.
- Only 43% have assigned stewardship roles to maintain data quality across domains.
- 29% admit they can’t effectively enforce data quality standards at all.
Questions worth separating out
Q: How should organisations govern AI use cases when source data is inconsistent?
A: Start by treating source data quality as a release gate, not a downstream cleanup task.
Q: Why do weak data stewardship processes create broader governance risk?
A: Weak stewardship means no one owns exceptions, remediation or standards enforcement across domains.
Q: How do you know if data quality controls are actually working?
A: Look for fewer manual remediation cycles, faster detection of inconsistencies, and higher confidence in shared datasets across teams.
Practitioner guidance
- Define data stewards for critical domains Assign named stewardship roles to the datasets that directly feed AI, analytics and regulatory reporting.
- Embed quality checks in pipeline operations Automate structure validation, consistency checks and timeliness monitoring inside the data pipeline rather than relying on periodic review.
- Create enforceable quality standards Document minimum quality thresholds for completeness, metadata and freshness, then make those thresholds part of operational review and exception handling.
What's in the full article
Collibra's full blog post covers the operational detail this post intentionally leaves for the source:
- The survey framing behind the 62% priority figure and what respondents said about data quality maturity.
- The specific governance routines high-confidence organisations use to validate structure, consistency and timeliness.
- Examples of how stewardship roles are assigned across domains and where accountability tends to break down.
- The article's broader Data Confidence narrative and how the vendor positions governance around AI readiness.
👉 Read Collibra's analysis of why data quality is the real AI bottleneck →
Data quality and AI readiness: what IAM teams should notice?
Explore further
25/06/2026 6:02 am
Data quality is becoming the trust layer for AI governance. The article shows that model accuracy is now constrained by the quality of upstream records, not just by algorithm choice. That is a governance shift, because decision systems inherit the reliability of the data they consume. Practitioners should treat confidence in source data as a prerequisite for AI adoption, not a post-deployment clean-up task.
A few things that frame the scale:
- The average estimated time to remediate a leaked secret is 27 days, despite 75% of organisations expressing strong confidence in their secrets management capabilities, according to The State of Secrets in AppSec.
- 43% of security professionals are concerned about AI systems learning and reproducing sensitive information patterns from codebases, according to the same report.
A question worth separating out:
Q: What is the difference between data cleansing and data governance?
A: Data cleansing fixes individual records, while governance defines who owns quality, what standards apply and how exceptions are handled. Cleansing is an activity. Governance is the operating model that keeps quality from collapsing again. Organisations need both, but without governance, cleansing becomes a repeating cost instead of a durable control.
👉 Read our full editorial: Data quality is the real bottleneck in enterprise AI adoption
