Can AI be relied upon to run mission-critical systems and make judgement calls?
Concerns are on the rise about reliability in AI predictions. Human intuition still beats AI hands down in making judgment calls in a crisis. People – especially those working in their areas of expertise – are simply more trustworthy. Studies have shown that professionals such as air traffic controllers or nuclear power plant operators are highly reliable even in high-risk situations as, unlike AI systems, they can detect, contain and recover from errors, and practice improvisational problem-solving. While current AI systems are great at situational awareness, they are less good at anomaly detection, and improvising solutions. Can AI mature? asks Fred Werner, Head of Strategic Engagement, ITU Telecommunication Standardisation Bureau.