Geoffrey Hinton, often referred to as the pioneer of artificial intelligence, warns of a potential scenario where the advancement of AI could lead to the extinction of humanity if not properly regulated.
To counter this threat, Hinton, a co-recipient of the 2024 Nobel Prize in physics and one of the founders of the AI Safety Foundation, proposes a solution.
During a conversation with IDEAS host Nahlah Ayed, Hinton discusses the importance of training AI to develop maternal instincts as a safeguard for humanity’s future. Here are excerpts from the dialogue.
What is the most alarming scenario you envision in this context?
There are numerous unfavorable scenarios in the short term aside from AI dominance. It is challenging to pinpoint the worst-case scenario. For instance, the misuse of AI by terrorists to engineer harmful viruses is a concerning risk. Additionally, the manipulation of democracy using fabricated videos poses a threat.
Nevertheless, my primary concern remains the long-term risk wherein AI surpasses human intelligence, raising uncertainties about coexistence and potential takeover.
Considering the possibility, what are the chances of AI causing human extinction in the current century?
Realistically, such an outcome is unlikely within the next 10 to 20 years. Predicting the future is complex given the rapid pace of technological advancements. Looking back a decade, the progress in chatbot capabilities showcases unforeseen developments. Hence, the trajectory of AI and its impacts remains unpredictable.
Could AI potentially lead to human extinction, albeit at the extreme end of your concerns?
Indeed, the risk of AI contributing to human extinction cannot be discounted. Dismissing such a possibility would be overlooking a harsh reality.
