Gartner: AI coding assistants to surge, but managing expectations is key
-
By 2028, Gartner estimates 75% of enterprise software developers will use AI coding assistants, up from <10% in early 2023.
-
However, Gartner warns expectations of productivity gains may be unrealistic - vendors claim 50% gains but that may not translate to overall dev cycle time.
-
Narrow AI coding tasks like boilerplate code don't represent overall software dev complexity.
-
Forcing adoption of AI coding tools top-down is a bad idea - teams need freedom to experiment.
-
CIOs should create a culture open to experimentation/failure to get the most benefit from AI coding tools.