Why 95% of AI Automation Projects Fail

Author(s): Luca Derumier

Originally published on Towards AI.

Why 95% of AI Automation Projects Fail

As CTO at Codika, I spend a lot of time talking to business leaders about automation. Lately, these conversations have followed a predictable pattern. They start with excitement — “2025 is the year of AI agents!” — and end with frustration: “We tried to automate X, but it went nowhere.”

The article discusses the alarming rate of failure in AI automation projects, primarily attributing the 95% failure rate to underlying issues rather than superficial challenges like data quality or integration complexities. It highlights a paradox where, despite the market’s recognition of AI’s potential, most companies struggle to convert automation ideas into successful implementations. The author emphasizes that the real challenge lies in the “Translation Barrier” where the original business requirements are diluted through layers of technical translations, ultimately hindering project success.

Read the full blog for free on Medium.

Join thousands of data leaders on the AI newsletter. Join over 80,000 subscribers and keep up to date with the latest developments in AI. From research to projects and ideas. If you are building an AI startup, an AI-related product, or a service, we invite you to consider becoming a sponsor.

Published via Towards AI