AI Mania Is Destroying Rational Decision-Making Across Organizations
Original: AI Mania Is Eviscerating Global Decision-Making
Why This Matters
A growing body of firsthand accounts challenges the narrative of enterprise AI ROI, raising questions about capital allocation and board-level accountability.
A consultant with 300+ professional catchups and 1.5 years of sales and technical engagement experience reports a 0% success rate across all observed AI projects in both private and public sector organizations, arguing that institutional 'AI psychosis' has made rational discourse impossible.
Software consultant and blogger 'Ludicity' published an essay on July 18, 2026 arguing that widespread AI adoption across corporations and government institutions has produced no measurable success. Drawing on over 300 catchups with professionals ranging from niche service industries to Fortune 500 executives, the author states that every single AI project observed over the past 18 months has failed — including projects observed only tangentially during unrelated work.
The author attributes the lack of honest reporting to structural incentives: executives who publicly criticize AI initiatives risk removal, employees risk termination, and vendors, consultants, and boards all benefit from obscuring failure rates. Many publicly traded companies, the author claims, have done nothing beyond purchasing Microsoft Copilot licenses yet are announcing 'AI productivity gains' to the press.
The essay quotes Mitchell Hashimoto, co-founder of HashiCorp and creator of Ghostty, who wrote: 'I strongly believe there are entire companies right now under heavy AI psychosis and it's impossible to have rational conversations with them about it.'
The author argues that failure is often not attributable to AI technology itself, but to companies' pre-existing inability to run software projects effectively. An essay editor noted that their division pivoted twice — first to build agentic workflow interfaces (used by only 10 users), then to agentic workflow support — finding a crowded market with few viable directions.