🚫AI cannot bring agility and efficiency across decision-making.
The ai, as everyone knows it today, i.e., the LLMs and GPTs, consumed data to learn how to make predictions. That's also true in the case of non-generative machine learning techniques.
So that's decision-making?
Well… the prediction values are N🚫T decision-making.
Predictions can be used in some parts of decision-making.
🚩The value of predictions, e.g., probability, classification, regression, and forecasting values, are just simple values that a machine spits out based on what it is trained for.
The same applies to the sentences you see LLMs and GPTs put on your screen by predicting the next words based on the previous one — still a prediction.
What to do with these prediction values?
✨Make predicted values part of the holistic decision model. That's where decision-making happens.
How?
💡The “decision model” is a graph that connects some components of decision-making, similar to neurons. Each of these nodes is related to a specific angle of the decision-making process.
* Some are based on businessrules and policies.
* Some are based on procedures and operation knowledge.
* Some are based on computational logic and formulations and SME knowledge.
* Last but not least, one of the nodes is based on the ML (generative and non-generative) prediction values.
The “decision model” brings these to life and makes the decision execution happen. It involves multiple different sets of techniques.
💣This approach is the foundation of agility and efficiency across decision-making in many organizations such as insurance, banking supplychain government health, and financialservices.
–
Follow me if you are looking for unfiltered views into decisionintelligence ai. Hit 🔔on my profile and get a notification for all my new daily posts.
Posted here.
Published December 17th, 2024 at 07:30 am
