The first study on the impact of artificial intelligence (AI) generative AI in a workspace has revealed that it improved productivity by 14%.
Bloomberg reports on a study conducted within a software in the list Fortune 500 whose name was not disclosed.
Researchers from the Stanford University and the Massachusetts Institute of Technology reviewed the impact of this type of tool over the course of a year.
Artificial intelligence at work: what was its impact?
The performance of more than 5 thousand customer service agents, based primarily in the Philippines, was tracked by researchers.
The agents were divided into two groups: one with access to tools of artificial intelligence trained based on data about good conversations with customers and another away from these tools.
The artificial intelligence succeeded in improving worker performance, having a greater effect on those workers who were less trained, getting them to perform their work 35% faster.
Overall, a 14% improvement in productivity was observed.
And what about those who train AI?
The study suggests that this “bump” in the performance of less skilled workers is achieved by the knowledge absorbed by the artificial intelligence of top performers and then disseminated to the rest of the company.
The researchers found that the IA gave little or no benefit to the more skilled workers, who rather fed the databases with their findings and reports.
The question remains as to whether “star workers” should receive extra compensation for training a IA which will ultimately end up benefiting the entire company.