Prof. Gianni / Rob&AI

Questions & Answers

How would you define Artificial Intelligence (AI)?
Artificial intelligence (AI) is concerned with building machines with the power to carry out operations that usually require human intelligence; that noted, this includes understanding verbal information, learning, thinking, and problem-solving.

AI has existed for nearly 70 years. Why is it experiencing impressive growth right now?
The main elements that have contributed to AI's recent explosion include: the presence of large amounts of training data, increased power in processing capacity (GPUs), and algorithmic advancements (neural networks).

Can you identify three domains or classes of problems where current AI models still face significant challenges?
Common-Sense Reasoning: AI is not able to naturally understand things like humans do because it does not have the ability to sense or feel.
Explainability: Since many models are "black boxes," it might be challenging to have faith in their judgments (Hassija et al., 2024).
Generalization: While models are excellent at specific tasks, they are unable to apply their expertise to things that are not researchable on the internet\.

Can you name at least three application sectors where robots are widely used, and explain why they are employed in those sectors?
Manufacturing: for the purpose of increasing productivity in repetitive processes (say, welding).
Logistics/Warehousing: To sort and move items daily in a way that minimizes expenses.
Surgery: To improve patient outcomes, more so through increasing a surgeon's precision.

Do you believe that the convergence of Artificial General Intelligence (AGI) and fully autonomous robotics is imminent and could potentially have a major impact on the world?
No, this convergence is not happening anytime soon; such is because AGI is still only a theoretical objective with no apparent route to success.

References

  1. Hassija, V., Chamola, V., Mahapatra, A., Singal, A., Goel, D., Huang, K., ... & Hussain, A. (2024). Interpreting black-box models: a review on explainable artificial intelligence. Cognitive Computation, 16(1), 45–74. Link
  2. Russell, S., & Norvig, P. (2020). Artificial Intelligence: A Modern Approach (4th ed.). Pearson. Book site