Eliezer Yudkowsky, who has spent 30 years on the AI safety problem, makes a firm prediction: if anyone builds a superintelligence using anything resembling current methods, everyone will die. He says Geoffrey Hinton, the Nobel laureate who invented deep learning and left Google specifically to speak freely, estimates a 25 to 50 percent chance of catastrophe and Yudkowsky considers that too optimistic.
He warns that today’s AI systems are already designing functional viruses from scratch, predicting protein folding, and that a superintelligence would simply connect those capabilities to build its own self-replicating biological infrastructure independent of humanity.
He compares the AI industry’s self-justification to leaded gasoline and cigarette companies, says capabilities are advancing orders of magnitude faster than alignment work, and describes the only historical precedent for stopping something like this: humanity chose not to use nuclear weapons, and that is the model.
📚 Sources cited in this video:
– Eliezer Yudkowsky, LessWrong – AI Alignment Writings
– Machine Intelligence Research Institute, Technical Research
– Geoffrey Hinton, Nobel Prize in Physics 2024 – Nobel Committee
– Future of Life Institute, Statement on AI Catastrophic Risk
⚠️ DISCLAIMER: This channel provides AI commentary and analysis for educational and informational purposes only. Views expressed by guests are their own and do not represent the positions of any company or institution. We encourage viewers to consult multiple sources and form their own conclusions.
#ai #agi #artificialintelligence
source
💸 Earn Instantly With This Task
No fees, no waiting — your earnings could be 1 click away.
Start Earning
