Cloudera has released the findings of its latest survey report, ‘The Future of Enterprise AI Agents’. The survey polled nearly 1,500 enterprise IT leaders across 14 countries to understand their adoption patterns, use cases, and sentiments around AI agents. Results show that an overwhelming 96% of respondents have plans to expand their use of AI agents in the next 12 months, with half aiming for significant, organization-wide expansion. The applications for this deployment include performance optimization bots (66%), security monitoring agents (63%), and development assistants (62%).
AI agents are autonomous software systems that can reason, plan, and act on behalf of users and when implemented effectively, they can unlock operational agility, drive cost savings, and dramatically improve customer engagement. As a result, AI agents are quickly becoming a key source of competitive advantage, with 83% of organizations stating that investing in them is crucial to maintaining their edge in the market.
However as with all emerging technologies there are risks to consider.
Agentic AI carries the risk of undermining consumer and employee trust, says the report. When AI is trained on historical data, it can reinforce societal biases unintentionally and influence outcomes.
Over half of respondents (51%) have significant concerns about AI bias and fairness. Bias is not confined to flawed data; it can manifest in
how workflows are structured, how intent is interpreted, and how outcomes are evaluated.
The report advises that to establish trust, organizations must prioritize data quality, ensure model robustness, and adopt ethical decision-making practices. This involves thoroughly testing AI models to eliminate biases, implementing strong data governance and security measures, and conducting regular audits to maintain trust throughout the AI system’s lifecycle.
To combat bias, enterprises are also taking additional steps to govern AI agents responsibly. Many respondents (38%) are instituting multiple processes, such as human review, diverse training data, and formal fairness audits. Another 36% have introduced some bias-check measures, such as periodic human reviews or bias-detection tools.
Concerns about data protection risks were also raised by survey respondents, with 65% of IT leaders looking for stronger data privacy and security features to be implemented in AI agents. 53% of overall respondents said that data privacy concerns were holding back AI agent adoption in their organization.
