🚩 “By some estimates, more than 80 percent of AI projects fail” – RAND
My guess is if you start with AI that's what you get!
💥The biggest challenge is to avoid the AI hype and excitement where that all the sudden everything is AI and organizations leaders #insurance #banking #financialservices #health #government have forgotten they are responsible for outcomes and, avoid getting caught in the hype of #machinelearning #ai #LLMs.
Start from basics – what does it look like to be outcome driven?
It means understand what outcomes you are trying to achieve NOT about:
❌ project requirements
❌ #AI algorithms #LLMs #MachineLearning
❌ #data governance and strategy
💡When you want to be outcome-driven, you are going to need to start with the outcome. They are almost always have strong association with “business decisions”. Outcomes are derived by the decisions organizations make and execute.
So what?
🎯Start “decomposing” the outcomes to “business decisions” that are made and execute in organizations which they influence the outcome.
You may ask Decomposition? How does that work?
🚀This means, let's put the outcome up and ask the series of questions as what decisions do we make which influences that outcome. And keep doing that to the rest of the decisions until you build the holistic view of the business decisions meaning the “decision graph” that determines how the outcomes are created.
Then what?
💣In the Decision Graph, you will have a set of decision units that are connected and influence each other where eventually they create the outcome. Which means you have the blueprint and roadmap on how to improve the outcome by improving those individual decisions in your Decision Graph! ✨No technology, #llms #ai #algorthim can automate this part for you. It needs humans, the domain experts and executives sitting together and figure out the DecisionGraph in collaboration. #noai
–
Follow me if you are looking for unfiltered views into #ai for decision-making #decisionintelligence. Hit 🔔on my profile and get a notification for all my new daily posts.
Posted here.
Published October 16th, 2024 at 07:30 am
