FICO Survey Spotlights Critical Explainability Gap Despite High AI ROI

A recent FICO‐commissioned study reveals a striking divide: while a majority of enterprises report strong returns from AI deployments, hardly any can confidently explain AI decisions to stakeholders. The findings spotlight a widening “explainability gap” that may hinder broader adoption of AI in critical decision-making processes.

Survey Results Underscore Explainability Deficit

FICO’s 2026 “State of Responsible AI” report—conducted with Corinium Global Intelligence—surveyed 1,004 technology, risk, and data executives. It found that only 5.2% of respondents said they feel very confident explaining AI-driven decisions to regulators or customers, while over one-third (37.3%) are neutral or not confident at all. Meanwhile, 85.1% say AI has met or exceeded their ROI expectations, creating a paradox between performance and transparency.

Despite heavy investment, 71.5% of customer-impacting decisions are still made without AI, indicating that generative and predictive systems aren’t yet achieving operational depth across organizations.

The findings signal that while enterprises are extracting value from AI, most lack the infrastructure to articulate how those value-generating decisions are made.

Governance, Skill Gaps, and Infrastructure Hinder Adoption

Survey respondents identified key areas needing improvement: model development standards (71.9%) and interpretable model architectures (67.9%) as the most impactful responsible AI practices. Other widely adopted tools include bias detection (58.8%) and regular audits (53.5%), but full AI governance remains elusive.

Only 8.2% of organizations have a fully implemented shared AI deployment platform, while another 71.2% report partial capabilities. Most enterprises remain in early exploration or pilot phases of agentic AI deployment, with security concerns (40.1%) and data integration or quality issues (38.3%) cited as primary barriers.

Why It Matters for Enterprise AI Strategy

The disparity between perceived ROI and explainability raises serious governance and regulatory risks. As executives push to scale AI technologies, the inability to explain decisions could become a critical liability in regulated industries.

The survey highlights a pivotal transition point: leadership must now move beyond value generation to building transparent, auditable AI systems. Investments in governance infrastructure, shared platforms, and interpretable architectures are no longer optional—they are strategic necessities.

Stock investors and industry watchers should monitor how FICO leverages its decision-intelligence platform to address this gap, and whether emerging regulatory frameworks accelerate enterprise demand for explainable AI solutions.