Opinion ·
Why small businesses' AI projects fail and how to avoid the pitfalls
- Publication
- Mexico Business News (Mexico)
- Author
- Unsigned
- Published on
- 22 September 2026 · read in the page structured data
Summary
Sebastián Romo, a consultant who specialises in supporting SMEs, analyses AI deployment failures in small businesses, starting from the case of a poorly configured chatbot. He cites the OECD (31% of SMEs using generative AI among 5,000 companies in seven countries), the US Chamber of Commerce with Teneo (68% of US small businesses use AI on an ad hoc basis), and RAND (80.3% of enterprise AI projects fail to deliver the promised value). Drawing on a census of 839 AI training programmes in Spanish, he points out that only 8.9% teach the professional judgement needed to evaluate AI outputs, compared with 44% focused on technical use. He cites Gartner, which forecasts that more than 40% of agentic AI projects will be cancelled by the end of 2027, and identifies five high-potential uses for small businesses: first-line customer service, operational content, internal documentation, administrative support, and standardised quotes.
Summary based on the full text of the article.
Read the article on Mexico Business News →Figures cited
-
31 %
share of SMEs using generative AI in the OECD survey
5,000+ SMEs in seven countries, 2025
« The Organization for Economic Co-operation and Development (OECD) surveyed more than 5,000 small and medium-sized businesses across seven countries in 2025: 31% use generative AI. »
Source: OCDE
-
68 %
share of US small businesses using AI on an ad hoc basis
US small businesses, 2025
« The US Chamber of Commerce, together with Teneo, reported that same year that 68% of American small businesses use AI, but mostly ad hoc, to draft an email or summarize a document, with no strategy and no usage policy. »
Source: US Chamber of Commerce et Teneo
-
80.3 %
share of enterprise AI projects that fail to deliver the promised value
enterprise AI projects
« RAND documented that 80.3% of enterprise AI projects fail to deliver the value promised. »
Source: RAND
-
8.9 %
share of AI training programmes in Spanish that teach professional judgement
839 AI training programmes in Spanish
« Only 8.9% teaches the professional judgment needed to evaluate whether what the machine produced is actually any good. »
-
44 %
share of AI training programmes focused on tool use
839 AI training programmes in Spanish
« Of that offering, 44% teaches how to use the tool: generate, automate, write prompts. »
-
40 %
share of agentic AI projects that will be cancelled by the end of 2027
agentic AI projects
« Gartner predicts more than 40% of agentic AI projects will be canceled by the end of 2027, and the causes it names don't include model capability: spiraling costs, unclear business value, inadequate risk controls. »
Source: Gartner
-
57.3 %
share of companies not using generative AI because it does not suit their type of work
OECD survey
« In the OECD survey, the top reason for not using generative AI wasn't fear or cost: 57.3% said it doesn't fit the kind of work their company does. »
Source: OCDE
Classification
- Country
- Mexico, United States
- Company size
- SMEs, Micro-businesses
- Industries
- All industries, or industry not specified
- Keywords
- ai project failure, small businesses, ai agents, ai training, customer service, chatbot
Entry written by Marsanne Studio from the cited article, with the help of an AI model (claude-sonnet-5). The rights to the article and its illustrations belong to Mexico Business News. Entry translated from French by Claude Sonnet 5.5.