The dire warnings about AI keep coming.
Eric Schmidt, former chief executive of Google, says that within a few years, millions of independent AI agents working together “will develop their own language.” And “we won’t understand what they’re doing.” They will have escaped human control.
The University of Montreal computer scientist and AI pioneer Yoshua Bengio told The Guardian this week that AI models were showing signs of self-preservation, the capacity to evade guardrails and to inflict harm on humans. Without controls, he said, AI systems will be free to operate like hostile extraterrestrials.
