Gartner’s AI Hype Cycle: GenAI and the Trough of Disillusionment

According to Gartner’s 2025 Hype Cycle for Artificial Intelligence, GenAI has officially begun its descent into the Trough of Disillusionment. EDiscovery founder Doug Austin reports that GenAI is entering a critical new phase in its evolution, signaling a moment of reckoning for the once-celebrated technology. 

Gartner’s Hype Cycle is a framework that helps organizations understand how technologies evolve and assess their readiness for adoption. It outlines five phases: Innovation Trigger, Peak of Inflated Expectations, Trough of Disillusionment, Slope of Enlightenment, and Plateau of Productivity.

After a year riding the Peak of Inflated Expectations, GenAI is now facing mounting skepticism as businesses struggle to demonstrate tangible value. Despite an average $1.9 million in GenAI investments during 2024, fewer than 30% of AI leaders say their CEOs are satisfied with returns. Organizations with low AI maturity face unrealistic expectations and struggle to identify useful applications, while more advanced firms cite difficulties in sourcing skilled professionals and embedding GenAI literacy into their operations.

Governance and regulatory concerns are adding further friction. Persistent issues such as AI hallucinations, fairness, and bias continue to plague deployments, while evolving government oversight may slow innovation in areas like productivity and job automation. These factors are contributing to a sobering reassessment of GenAI’s immediate business utility.

The Hype Cycle model cautions against alarmism. Most emerging technologies descend into disillusionment before maturing. The key variable is how long GenAI remains in this trough before ascending the Slope of Enlightenment, when second-generation tools and real-world benefits begin to emerge.

Meanwhile, not all AI trends are dimming. AI-ready data and AI agents are the fastest-rising stars on this year’s Hype Cycle, both sitting at the Peak of Inflated Expectations. Agentic AI, in particular, continues to dominate discussions, reinforcing the broader momentum within the AI ecosystem even as GenAI regroups.