The new report “Genai Gap: The Status of Artificial Ielligence in Business 2” by the University of MIT shows that only 2 % of the productive artificial ielligence tests have achieved rapid growth in revenue; While most projects have stopped or have no significa economic impact.
This research is based on 1 Ierview with Managers, Survey of 5 Employees and Investigating 2 General Artificial Ielligence The carried out a big gap between limited success stories and broad organizational failures.
Why do most projects fail?
According to the saying Aditia ChalapaliThe main author of the MIT report is not the main problem of the quality of the models, but the organizational learning gap and the incomplete iegration of tools with iernal processes. He explains that tools like ChatGPT are excelle for individual use, but in the organizational environme they stop because of the lack of compatibility with the workflows.
In corast, smaller startups have been able to make tens of millions of dollars by focusing on a specific issue and fast implemeation.
The allocation of wrong resources
More than 1 % of productive artificial ielligence budget Is spe on sales and marketing in companies; While the findings of MIT indicate the highest return on capital in Office automation And removal is processes outsourcing.
Buy or build?
According to data, buying specialized tools and collaborating with vendors, in 1 % Has been successful; But domestic developme has only been concluded in one -third. This gap is particularly more common in setting industries such as financial services; Where companies have experienced a high failure rate despite heavy costs for domestic construction.
Consequences of labor force
The effect of artificial ielligence on the labor market is also evide. Many companies do not fill vacancies instead of widespread expulsion. Most changes have taken place in roles that were previously outsourced, such as customer support and office jobs.
Look at the future
The MIT research also pois to the growth of “shadow artificial ielligence” in organizations and the difficulty of measuring productivity. At the same time, the most advanced companies toward Factor artificial ielligence Move; Systems that can learn, remember and act independely. These technologies show a picture of the next stage of organizational artificial ielligence.
