4-model average
building category history
loading
8pt spread
The three supporting readings tell you how much weight to put on the probability: confidence reflects category-level track record, stability tracks how the estimate has moved over time, models shows whether the four agree.
Loading belief trajectory...
Given the current landscape, there is significant skepticism regarding the long-term viability and practicality of generative AI technologies as organizations navigate issues like data privacy, hallucinations, and regulatory scrutiny. Recent reports from OpenAI and Meta highlight mounting concerns among users, reflecting a potential shift from enthusiasm to disappointment. Historical trends show that technologies often face backlash after an initial hype phase, with a notable precedent seen in the AI hype cycles from the past.
Generative AI is currently experiencing peak hype, evidenced by massive investment and widespread adoption across industries, characteristic of the 'Peak of Inflated Expectations'. However, the rapid emergence of ethical concerns, regulatory scrutiny (e.g., EU AI Act nearing implementation), and the inherent limitations of current models in areas like accuracy and bias suggest a probable downturn towards the 'Trough of Disillusionment' by 2026, as early optimism gives way to pragmatic evaluation of challenges.
Gartner's Hype Cycle follows predictable patterns where technologies peak within 5-10 years then decline into the Trough. Generative AI peaked in hype in 2023-2024 (ChatGPT's explosive adoption, massive enterprise investments, regulatory uncertainty). By 2026 (2-3 years from peak), the technology typically enters the Trough phase as early implementations reveal limitations, ROI underperformance, and integration challenges become apparent. Current signals support this trajectory: enterprise GenAI adoption has already plateaued from 2024 peaks, multiple companies have scaled back AI investments (IBM, Meta reducing headcount), and disillusionment narratives are accelerating. However, there's meaningful probability (38%) it remains on the "Peak of Inflated Expectations" longer if a major breakthrough occurs (AGI progress, killer applications) or if Gartner's positioning diverges from typical patterns.
Gartner placed generative AI at the Peak of Inflated Expectations in its August 2023 Hype Cycle, and the typical 18-30 month progression from Peak to Trough across 40+ prior technologies indicates a 2025-2026 window; enterprise surveys show AI project ROI falling from 45% in 2023 McKinsey data to 28% in 2024 Deloitte benchmarks, while 2024 capex guidance from Microsoft, Google, and Meta already shows sequential growth slowing from 45% YoY to projected 22% YoY by 2025, consistent with capital reallocation that historically precedes Trough placement.