The 3D Decisions-Powered Enterprise Strategy for AI

In the 9th edition of this newsletter entitled “Enterprise Generative vs. Discriminative AI: The Need for the Balanced Strategy ”, it was concluded that enterprises shouldn’t design their “AI Strategy” to depend on one type of AI system solely or solely due to its foundational category and shouldn't favor one type over the other based on the current hype. However, the “Enterprise AI Strategy” should be designed based on a deep understanding of the “Enterprise’s Environment and its Critical Success Factors” as well as using the AI systems to build core competencies that can match these critical success factors and capture value from them. This strategy should strike the optimum balance between the use of “Generative and Discriminative AI Models”. This specific strategy is going to be the first dimension of the “Enterprise Strategy for Artificial Intelligence”.

Furthermore, in the 12th edition of this newsletter entitled “Enterprise AI Monetization: The Intelligent Strategy for Artificial Intelligence ”, it was concluded that enterprises shouldn’t design their “AI Strategy” from a technical perspective solely or solely due to its foundational benefits and shouldn't favor a certain AI model over another based on its technical function. However, the “Enterprise AI Strategy” should be designed taking into consideration the deep understanding of the enterprise’s environment and its “Economic Dynamics” and be extended into an “Enterprise AI Monetization Strategy” that should positively impact the “Tangible” or “Intangible” assets. This specific strategy is going to be the second dimension of the “Enterprise Strategy for Artificial Intelligence”.

However, this edition of this newsletter will introduce the third dimension of the “Enterprise Strategy for Artificial Intelligence”, which is the “Role” that the AI model is going to play in the “Enterprise Operational Model”. This AI model is either going to assist the performer of certain tasks or augment the performer of certain tasks. Simply, it is either an “Assisting Intelligence Model” or an “Augmented Intelligence Model”.

Being considered the most fundamental level of artificial intelligence, “Assisted Intelligence” is largely used to automate simple procedures tasks, and activities within the enterprise’s operational model. In this type of AI, AI technologies assist humans in doing their regular jobs more effectively or precisely. In this situation, machines are simply a tool that is operated under the supervision of normal human operators. For example, the car driver usually uses the intelligent GPS system during his journey to reach his required destination more effectively and make navigation decisions precisely.

On the other hand, “Augmented Intelligence” enables organizations and individuals to perform tasks, and activities they could not do otherwise by helping humans by transferring the full task or activity to the machine rather than just trying to imitate human intelligence. In this situation, machines aren’t just a tool that is operated under the supervision of normal human operators but a tool that complement the normal human operators. For example, the car driver usually uses the cruise control system during his journey to reach his required destination more effectively and make speed-limit decisions precisely in a way that exceeds his normal abilities.

Hence, by combining the three dimensions together, the result will be a “3D Decision” about each “Value-added AI Model” used by the enterprise, with “Eight Possible Combinations” in total.

For example, in one decision point, the enterprise can choose to build a value-added AI model that is discriminative by nature (i.e., the 1st dimension), which is able to identify the most profitable cluster of customers that can be targeted by the new advertising campaign. The ultimate target of this model is to increase sales revenue by a significant percentage and hence it aims to positively affect the tangible assets of the enterprise within its AI monetization strategy (i.e., the 2nd dimension). Furthermore, it is evident that this model is going to augment human intelligence as it was so difficult for the enterprise to identify this cluster of customers in the first place (i.e., the 3rd dimension).

On the other hand, in another decision point, the enterprise can choose to build a value-added AI model that is generative by nature (i.e., the 1st dimension), which is able to support its call center team of agents. The ultimate target of this model is to increase customer satisfaction levels by a significant percentage and hence it aims to positively affect the intangible assets of the enterprise within its AI monetization strategy (i.e., the 2nd dimension). Furthermore, it is evident that this model is going to assist human intelligence as the enterprise was able to support its customers in the first place, but this model is going to help the enterprise do this more effectively and precisely (i.e., the 3rd dimension).

Hence, to conclude, the “Enterprise AI Strategy” in one of its simplest images, can be seen as a “Series of Decision Points”. Each one of them can be powered by a “3-Dimension Decision” that should be taken or selected from the available “Eight Possible Options” for the “Value-added AI Models”. Of course, the collection of these 3D decisions is going to shape the overall power of the enterprise strategy for AI.