Building Trustworthy AI Starts with Data Confidence, According to Protiviti Study

MENLO PARK, Calif., Aug. 19, 2025- AI success is built on strong data, with 69% of organizations that have high AI maturity saying they are "very confident" in their data capabilities. Effective governance, paired with data engineering, analytics and training people to use data responsibly, is a key lever in advancing AI maturity and improving the return on AI investment. That is a key finding from global consulting firm Protiviti's second report in its AI Pulse Survey series, titled From Data Confusion to AI Confidence, which explores the relationship between confidence in data quality and its impact on both AI maturity and ROI.

As organizations mature, their data practices become more structured and intentional. At stage 5, the highest level of AI maturity:

• 74% of organizations conduct regular data audits - more than double the rate at stage 1 (36%).
• 57% have implemented robust data management policies and standards, compared to just 24% at the starting point.

Among organizations in which ROI on AI investments has significantly exceeded expectations, nearly all (97%) are confident in their ability to obtain, organize and understand the data required to support their AI goals.

Surveying more than 800 respondents, including C-suite executives, board members and leaders across global industries, the report builds on Protiviti's initial AI Pulse Survey to reveal how a strong data foundation is critical to advancing AI capabilities and realizing measurable returns.

The survey examines organizations' AI maturity through five phases:

• Stage 1: Initial - Organization recognizes AI's potential benefits but has limited understanding and no strategic initiatives. KPIs have not yet been defined.
• Stage 2: Experimentation - Organization has initiated small-scale AI projects and pilot programs to assess feasibility and benefits.
• Stage 3: Defined - Organization has integrated AI solutions into existing business processes, enhancing operational efficiency and decision-making.
• Stage 4: Optimization - Organization has optimized AI systems for performance and scalability, with continuous improvements based on data feedback.
• Stage 5: Transformation - AI drives significant business transformation at the organization, creating new opportunities and reshaping the industry landscape.

Expert Perspective

"Data confidence should be a primary focus for any AI strategy," said Peter Mottram, global leader of Protiviti's Enterprise Data and Analytics practice and a leader of its AI practice. "Organizations that trust their data are more likely to exceed expectations, scale effectively, and innovate with confidence.

"However, waiting for perfect data can stall progress," Mottram added. "Even messy data has value, and AI can help improve it. Organizations that invest in data literacy and governance will move more confidently along an AI maturity curve."

Key Findings

Where Data Confidence Leads, ROI Follows

Organizations with high data confidence are more likely to exceed AI investment expectations:

• Stage 2: Organizations begin piloting AI projects. Confidence levels start to improve, and there's a noticeable shift toward slightly exceeded expectations among those who are confident or very confident in their data.
• Stage 5: 69% are "very confident" and 76% overwhelmingly report exceeding ROI expectations.

Insight: Data maturity underpins AI value. Trustworthy, well-governed data enables scale, reduces bias, and delivers meaningful outcomes.

Barriers to AI Performance

Notwithstanding the level of maturity, data challenges persist:

• 57% of AI-mature organizations cite inadequate data governance as a significant challenge.
• Even mature companies struggle with security and compliance oversight.

Confidence by Sector, Region, and Role

• Industry: Tech and financial services are among the leaders in data trust; manufacturing and distribution lags behind.
• Geography: Respondents from countries that have invested in data standardization and quality programs have greater confidence in their data.
• Role: Confidence increases with seniority, peaking at the C-suite level.

Identification and Addressing Bias Risk Evolves with Maturity

• Stage 1: 30% report no bias-likely due to limited detection or simpler use cases where bias is less likely to occur.
• Stages 2-3: Bias is most frequently reported, as detection improves but mitigation tools are still developing.
• Stage 5: 28% report no bias-reflecting stronger mitigation strategies that identify and address issues before they affect results.

Maturity Includes Greater Awareness of Challenges

• The most mature companies have the experience to anticipate challenges and to be prepared for them.
• Stage 5: 59% cite security and compliance as top concerns (up from 41% at Stage 1), because maturity brings awareness of potential challenges and the strategies to address them.

Methodology

Protiviti surveyed over 800 global professionals-including C-suite executives, board members, and managers-across industries such as technology, finance, and operations. The AI Pulse Survey series, launched in 2025, explores themes like AI readiness, ROI, data challenges, and governance risks.