It’s been about a year since major headlines spoke of an insider at Google that felt that Artificial Intelligence (AI) had become ‘sentient’. Fast forward to today where AI is everywhere and has been stirring conversations ever since. It should be clear that while AI is a potential threat, it is ultimately a tool that can help us change our world for the better. Among the most existential conversations, a discourse has emerged surrounding AI’s potential dangers, distinct from other powerful technologies like biotechnology and nuclear weapons. While the latter technologies can cause devastation, they remain largely within our control; their catastrophic effects are typically the result of deliberate human decisions or misuses. The peril of AI, however, lies in the possibility it may one day escape human control entirely. This is not a difficult scenario to imagine as AI evolves at an ever-increasing pace. As AI becomes more capable, it’s possible to envision escalations or self-healing based on learning, which position the AI to seek control or dominance over its own environment and the systems under its control. This subtle mission could potentially iterate thousands of times before anyone notices, and in a moment, could become the prime directive for the existence of AI altogether.
Full opinion : Staying Ahead Of The AI Curve: The Imperative Of Prudent Planning.