Seeing the entire system and moving parts allows for a holistic approach in pursuing projects, especially AI implementation ones. I am currently in the midde of one :)
Of course. For example, the current project I am on, myself and one of the key team member are deeply thinking through the impacts to our collaborator teams and weighing the risks of lack of adoption.
Great advice. When people ask me how to implement AI in their business, I always tell them to start with documentation. If you have a process that you want to automate, document every part of it in excruciating detail - not just the steps of the process itself, but also the context around why it's being done, who's involved, what weird edge cases exist and what the output is being used for. Write documentation as if you're going to hand it off to a really smart employee who will read and absorb everything you put down on paper. At some point, you'll be able to feed that to an AI agent who can just take the whole thing over, but for now you can use it to break the process into discrete steps and figure out which ones are manageable via AI.
Seeing the entire system and moving parts allows for a holistic approach in pursuing projects, especially AI implementation ones. I am currently in the midde of one :)
Share your insights with us, Colette.
Of course. For example, the current project I am on, myself and one of the key team member are deeply thinking through the impacts to our collaborator teams and weighing the risks of lack of adoption.
Agreed, Kamil.
Systems thinking actually becomes data that can be used to train AI models and gain an edge in the short and long-term.
In other words, testing hypotheses, documenting successes, and turning them into processes is the moat for companies today.
And especially in times where we still need humans as the connecting bridge.
Agreed. I could be going on a limb here, but I think we will always need humans as the connecting bridge.
I sure hope so
Great advice. When people ask me how to implement AI in their business, I always tell them to start with documentation. If you have a process that you want to automate, document every part of it in excruciating detail - not just the steps of the process itself, but also the context around why it's being done, who's involved, what weird edge cases exist and what the output is being used for. Write documentation as if you're going to hand it off to a really smart employee who will read and absorb everything you put down on paper. At some point, you'll be able to feed that to an AI agent who can just take the whole thing over, but for now you can use it to break the process into discrete steps and figure out which ones are manageable via AI.