Employees Doubt AI Reliability But See It as Essential for Work

According to Asana's Global State of AI at Work 2025 report, 64% of employees believe AI agents are unreliable, yet 76% view them as a "fundamental" part of the future of work.

Key Findings

• Adoption vs. Trust Gap: 74% of UK workers are already using AI agents, but nearly two-thirds worry about their reliability.
• Accountability Issues: 39% believe no one is responsible for AI mistakes, while 20% blame end users, 18% IT teams, and 9% the creators.
• Lack of Governance: Only 10% of organizations have clear ethical frameworks or deployment processes for AI agents.

The Growing AI Debt

• 79% of organizations risk accumulating "AI debt" due to unreliable systems and poor oversight.
• Only 18% of businesses measure AI errors, despite 63% of employees citing accuracy as a top priority.

The Need for Training and Clarity

• 82% of employees say training is "essential," yet only 32% of organizations provide it.
• Mark Hoffman, Work Innovation Lead at Asana, emphasizes: "Access to AI tools isn’t enough. Employees need training, clarity, and guardrails."

HR's Role in Bridging the Gap

Hoffman concludes that HR leaders are key to closing the trust gap by redesigning workflows and implementing clear governance. Organizations investing in these areas are already seeing real impact.

For more insights, read the full report here.