Former Google CEO Eric Schmidt has warned that artificial intelligence (AI) is vulnerable to hacking and could even be used to commit murder.
On Thursday, CNBC reported that Schmidt, speaking at the Sifted Summit – an IT forum – expressed grave concerns about the potential misuse of AI technology, especially the risk of AI systems being hacked to carry out killings.
Schmidt, who served as Google’s CEO from 2001 to 2011, addressed the question of whether AI poses a greater threat than nuclear weapons. He responded that AI could present similar proliferation challenges and might be exploited by dangerous individuals or groups.
“Learning how to kill someone is one example,” Schmidt stated, referring to the potential misuse during AI model training.
Schmidt emphasized the current lack of a “non-proliferation regime” to manage AI risks.
He also noted that AI is not receiving the level of attention it warrants given its potential impact.
Addressing recent speculation about an AI bubble, Schmidt dismissed comparisons to the dot-com crash. However, he added that he is not a professional investor, tempering his market predictions.
