{"id":170284,"date":"2026-07-17T14:26:01","date_gmt":"2026-07-17T13:26:01","guid":{"rendered":"https:\/\/www.intelligentcio.com\/eu\/?p=170284"},"modified":"2026-07-17T14:26:02","modified_gmt":"2026-07-17T13:26:02","slug":"the-ai-confidence-paradox-how-can-organisations-truly-trust-their-data-to-scale-agentic-ai","status":"publish","type":"post","link":"https:\/\/www.intelligentcio.com\/eu\/2026\/07\/17\/the-ai-confidence-paradox-how-can-organisations-truly-trust-their-data-to-scale-agentic-ai\/","title":{"rendered":"The AI confidence paradox \u2013 how can organisations truly trust their data to scale agentic AI?"},"content":{"rendered":"\n

Dave Shuman, Chief Data Officer, Precisely, says that trusted, high-integrity data is the essential foundation for organisations seeking to scale agentic AI successfully while reducing operational, regulatory and reputational risks.<\/em><\/p>\n\n\n\n

The global agentic AI market is estimated to reach \u00a325.9 billion in 2030 – up from a projected \u00a36.3 billion in 2026.<\/p>\n\n\n\n

Despite this investment, Gartner predicts that more than 40% of agentic AI initiatives will be cancelled before 2028.<\/p>\n\n\n\n

While many leaders rush to deploy and reap the benefits of autonomous agents \u2013 due to their ability to respond to real-time customer needs, manage operational workflows or coordinate multi-step business processes \u2013 they are failing to achieve a positive return on investment (ROI). This diminishing ROI is largely due to implementation, rather than the technology itself.<\/p>\n\n\n\n

It must be understood that AI is a product of the data that fuels it, and as organisations shift from pilots to autonomous, agent-driven systems, the gaps in data readiness are becoming impossible to ignore. In fact, this readiness gap can even cause serious business disruption and regulatory consequences. To truly achieve success in agentic AI initiatives, organisations must first consider whether they can trust their data.<\/p>\n\n\n\n

Trusted data for AI readiness – and what this means in the agentic era<\/strong><\/p>\n\n\n\n

Trust in the context of agentic AI is achieved when companies are confident that their data will consistently deliver reliable insights and drive sound decision-making. As opposed to LLMs, which generate text-based outputs in response to prompts, AI agents act autonomously within user-defined parameters.<\/p>\n\n\n\n

This shift towards agentic AI marks a decline in human oversight and a move towards autonomous systems that execute tasks independently. Agentic AI therefore demands data integrity: meaning data that is not just available, but complete, accurate, contextual and, ultimately, ready for real-time decision-making at scale. This is because, without trust, organisations risk automating uncertainty instead of insight. Autonomy combined with guardrails, transparency and flexibility is therefore crucial in ensuring agents are acting on data that can be trusted.<\/p>\n\n\n\n

Despite the importance of trusted data in the agentic era, the 2026 State of Data Integrity and AI Readiness<\/em> report found that 67% of data and analytics professionals do not fully trust the data that their organisation is using. Even with this lack of confidence, 32% expected to see a positive ROI from AI initiatives within the next 6-11 months.<\/p>\n\n\n\n

The disconnect between confidence and reality needs addressing, as without organisations revisiting their data frameworks and strategies to ensure accuracy, consistency and context, agentic AI success is far from guaranteed. Instead, implementation may create significant risks.<\/p>\n\n\n\n

The risks of low trust in the context of agentic AI<\/strong><\/p>\n\n\n\n

Additionally, without trustworthy, well-governed data, organisations cannot safely grant agents decision-making authority or autonomy. In fact, models trained on flawed, inaccurate or unrepresentative data will likely produce outputs that reflect those same deficiencies – a risk that holds true even for agents using frontier models, which are only as reliable as the data they utilise.<\/p>\n\n\n\n

In parallel with the rise of agentic AI, there is a growing demand for accountability and transparency, with recent findings revealing that only 14% of UK consumers are comfortable with fully autonomous AI.<\/p>\n\n\n\n

This lack of confidence is not without reason. Although agentic AI has huge potential, it can still make mistakes when organisations neglect data integrity principles including data governance, integration, enrichment and geospatial insights.<\/p>\n\n\n\n

For example, studies in recent years have found that, while agents work more quickly than humans, this productivity is often offset by lower accuracy \u2013 which stems from data. These inaccuracies are a cause for concern, as incorrect autonomous decision-making without effective guardrails in place can have serious real-world consequences across industries, leading to reputational damage, financial losses and non-compliance.<\/p>\n\n\n\n

Beyond reputational damage, for some industries including healthcare and financial services, errors caused by incorrect or outdated data could be life changing. For instance, a financial agent trained on unrepresentative or outdated data with the autonomy to approve or deny credit loans might consequently discriminate in its decisions, unbeknownst to the financial institution it serves.<\/p>\n\n\n\n

The EU AI Act’s maximum penalty for the most serious violations reaches up to \u00a330.5m in accordance with the EU AI Act, as well as increased scrutiny over how AI-driven decisions have been made.<\/p>\n\n\n\n

Similarly, agentic AI shows promise in monitoring patient data streams from wearable devices and autonomously adjusting treatment plans. However, if data is inaccurately recorded, without data enrichment to verify this against other health indicators, the agent might autonomously trigger inaccurate medical recommendations.<\/p>\n\n\n\n

A necessary industry shift from \u2018AI first\u2019 to \u2018data first\u2019<\/strong><\/p>\n\n\n\n

For organisations to ensure agentic AI readiness and mitigate these risks, they must therefore shift away from an \u2018AI first\u2019 approach and, instead, prioritise data from the outset. The data required to power agentic AI is often scattered across hybrid and legacy systems, incomplete or outdated, lacking context, non-compliant and expensive to manage.<\/p>\n\n\n\n

This creates blind spots and makes it difficult for systems to make accurate, autonomous decisions.<\/p>\n\n\n\n

Trust is a manufactured outcome for agentic initiatives; it does not simply occur once the technology is implemented. For example, organisations with a formal data strategy report 70% trust levels in their data, compared with just 50% for those without these strategies in place.<\/p>\n\n\n\n

Evidently, implementing a robust data integrity framework allows businesses to transform their data into an agentic-ready business advantage. Ensuring that strategies are in place prior to the implementation of agents means that when AI works autonomously, it does so on a foundation that can be trusted.<\/p>\n\n\n\n

Data integrity is the foundation to secure trust in agentic AI<\/strong><\/p>\n\n\n\n

Organisations must therefore strengthen foundational data practices to enable more effective, safer AI with better business outcomes. Agentic AI initiatives must be powered by high-integrity data to produce trustworthy, meaningful and representative outputs.<\/p>\n\n\n\n

Achieving this standard requires breaking down silos, ensuring data quality, enforcing rigorous governance and enriching training data with curated, AI-ready attributes and spatial insights.<\/p>\n\n\n\n

Governance must first be prioritised as a key driver of trust. Data governance programmes should be expanded to specifically include AI and must embed transparency and fairness at every stage to ensure reliability, quality and value. Those with governance programmes in place will see greater value across business outcomes, including operational efficiency, compliance, cost reduction and revenue generation.<\/p>\n\n\n\n

Data should also be implemented across hybrid and cloud environments. Siloed data is a critical barrier to agentic AI success, as when data is siloed across platforms, it is challenging for organisations to create an accurate view of the entirety of the information available to them. Integrating data across environments offers a more comprehensive view of data, consequently increasing trust in autonomous decision-making.<\/p>\n\n\n\n

A contextualised foundation is another key requirement for agentic AI. This should contain enriched first-party data combined with curated third-party sources \u2013 including environmental risk indicators, precise address data and demographic profiles. These insights allow for a broader understanding of exactly how autonomous decisions are being made.<\/p>\n\n\n\n

Finally, organisations must prioritise data quality and observability practices. They should ensure data accuracy, consistency and completeness to avoid silently introducing inaccuracies into agentic models. Continuously monitoring data for anomalies will also allow teams to proactively identify and address issues before they result in inaccuracies down the line.<\/p>\n\n\n\n

Agentic AI success begins with trust<\/strong><\/p>\n\n\n\n

By shifting the focus from the intelligence of the model to the integrity of the data, organisations can move past the agentic AI hype and into real, measurable business success. The question is no longer whether data is accurate, but whether it is ready to support real-time, autonomous decision-making.<\/p>\n\n\n\n

Agentic AI ultimately raises the bar for trust. Organisations must know exactly where their data comes from, how it’s being shaped and whether it can be used with limited intervention. Without that foundation, AI’s potential is either limited or may open the door to unacceptable risk.<\/p>\n","protected":false},"excerpt":{"rendered":"

Dave Shuman, Chief Data Officer, Precisely, says that trusted, high-integrity data is the essential foundation for organisations seeking to scale agentic AI successfully while reducing operational, regulatory and reputational risks. The global agentic AI market is estimated to reach \u00a325.9 billion in 2030 – up from a projected \u00a36.3 billion in 2026. Despite this investment, […]<\/p>\n","protected":false},"author":58,"featured_media":170285,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"footnotes":""},"categories":[21064,20673,93],"tags":[22048,158,19032,22362,22734,444,2594,20582,76,17618,25457],"class_list":["post-170284","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-ai","category-expert-opinion","category-top-stories","tag-agentic-ai","tag-ai","tag-ai-readiness","tag-ai-regulation","tag-autonomous-ai","tag-data-governance","tag-data-integrity","tag-data-quality","tag-digital-transformation","tag-precisely","tag-trusted-data"],"acf":[],"publishpress_future_workflow_manual_trigger":{"enabledWorkflows":[]},"_links":{"self":[{"href":"https:\/\/www.intelligentcio.com\/eu\/wp-json\/wp\/v2\/posts\/170284","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.intelligentcio.com\/eu\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.intelligentcio.com\/eu\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.intelligentcio.com\/eu\/wp-json\/wp\/v2\/users\/58"}],"replies":[{"embeddable":true,"href":"https:\/\/www.intelligentcio.com\/eu\/wp-json\/wp\/v2\/comments?post=170284"}],"version-history":[{"count":1,"href":"https:\/\/www.intelligentcio.com\/eu\/wp-json\/wp\/v2\/posts\/170284\/revisions"}],"predecessor-version":[{"id":170286,"href":"https:\/\/www.intelligentcio.com\/eu\/wp-json\/wp\/v2\/posts\/170284\/revisions\/170286"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.intelligentcio.com\/eu\/wp-json\/wp\/v2\/media\/170285"}],"wp:attachment":[{"href":"https:\/\/www.intelligentcio.com\/eu\/wp-json\/wp\/v2\/media?parent=170284"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.intelligentcio.com\/eu\/wp-json\/wp\/v2\/categories?post=170284"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.intelligentcio.com\/eu\/wp-json\/wp\/v2\/tags?post=170284"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}