Understanding AI in the context of your own business, industry, application area, technology evolution is the foundation for any AI implementation in the enterprise.

• Do not emulate or try to compete with the hyperscalars on AI. The market opportunity in AI is sufficiently bigger than what the hyperscalars are aiming for.
• AI is an umbrella of different technologies (machine learning, deep learning, computer vision, NLP, machine reasoning) and depending on the specific industry and application, there is likely to be a unique technology combination.
• AI application development is moving towards mobile application development with “drag and drop” interfaces, however enterprises should try and avoid “black box” approaches to “drag and drop” modules.
• AI ownership within enterprises is most likely better off with a decentralized approach, with CEOs in the long run becoming the default Chief AI Officers.
• Follow the right mix of “in-house + outsource” approach to implementations, based on circumstances.
• Data is fuel for AI, and as an investment priority should be on top of the list.