I have heard it several times now, folks in charge of API governance sharing with me that they are finding success in injecting API governance into developers workflow using artificial intelligence. They had tried changing behavior with teams through meetings, sprints, workshops, and other approaches unsuccessfully, and now they are finding that developers question things less when it is introduced via artificial intelligence available in their integrated development environment (IDE). It is one of those aspects of enterprise operations that seem like a positive technological solution, when in reality it is a sign of larger systemic issues in how people work together, communicate, and respect each other in the workplace.
A command line and chat prompt can provide a powerful way to abstract away humans, and removing the authoritative figure from governance, while “tricking” developers into changing their behaviors and following the rules might seem like a positive move, but it really isn’t any way to govern people, the enterprise, or any other state. The cultural issues that push engineering teams to disregard and disrespect API governance policies and rules won’t change unless teams are well informed, given agency, and are also expected to be accountable for their work. If you are looking to just obfuscate this, you are choosing to kick the API governance can down the road, which I am guessing has already been done several times before, resulting in you being at the intersection you are today.
It is painful to watch enterprises make such diminishing decisions regarding how they operate their APIs in the name of efficiency, cost savings, and trends. I am sorry that teams do not want to listen to API governance. It is hard to do governance in a way that is inclusive to teams, and to provide them with the proper guidance, but it is worth it in the long term. There are few shortcuts with API governance, and it is something that is more social than it is technical. While you may see some short term gains from injecting API governance into the AI prompt for your teams, you will also be missing out on a critical feedback loop with teams which is necessary to evolve your API governance and keeping it matching the reality on the ground. Rather than tricking your developers into listening using AI, I’d recommend incentivizing your developers to lead the way when it comes to API governance, finding what will motivate them to not just care, but help define and lead when it comes to governing operations.
