1. Where is the market in terms of real adoption of ML and AI?

There are many practical areas … however, in the deployment of AI and ML, it is important to be realistic, and more importantly, to differentiate between Sci-fi movies and deployment in reality. There are three major factors to consider : Work processes, work competency, workplace configuration. In particular, there is huge potential in the human-machine collaboration area. AI must augment human capability, and ML must enhance human capability, all in all, adoption must dovetail a varying set of human adoption, and not a one-size fit all.

2. What are the data issues executives need to solve – or at least understand – before considering AI adoption?

Data is not everything. I hesitate to endorse that past data can solve all future problems. Data analytics can demonstrate past trends, but can be biased around how and what that data sets represent. There will always be hidden data which is not captured in digital form but reside in the human thoughts. Hence, the human validation on an on-going basis is critical.

3. Challenges: What are the unstated assumptions that cause projects to get tripped up?

AI projects must be considered globally and not scoped out narrowly to work in silos. Adopting AI to “plaster” a problem is probably set to fail.

4. Plenty of AI projects end up as expensive science projects that businesses spend lavishly on, only to realize disappointment”. Do you agree with this statement? Why or why not?

This is true! This is because, in designing an AI solution, there are many considerations. It involves a transformation of work process, competency and environment. Scope creeps appear because the AI adoption is not considered in full spectrum.

– Oliver Tian – Honorary Advisor, Asia Pacific Assistive Robotics Association; Vice-Chairman, International Alliance of Robotics Associations