Google DeepMind researcher Matthias Bastian resigned, warning that building superintelligent AI soon is 'inherently irresponsible.' He cited concerns over AI's rapid development and its potential risks. Bastian, a technical expert at Google DeepMind in New Zealand, worked on chip design tools that made AI cheaper and faster, a contribution he now says he can no longer justify.
Bastian called the risk from superintelligent AI 'real but uncertain,' and argued that aiming for artificial superintelligence (ASI) in the near future is inherently irresponsible. He warned of cognitive surrender, AI-induced psychosis, and loneliness, as well as concentrated power, economic disruption, cybersecurity risks, and lack of accountability.
"The current rate of change is far too high," said Bastian. He also expressed concern that AI's harms may outweigh its benefits. Many DeepMind colleagues share his concerns but rarely speak publicly, he said.
He now wants to focus on 'unambiguously pro-human' work.
The announcement follows a broader debate within the AI community about the ethical implications of developing superintelligent systems. Bastian linked to 'If Anyone Builds It' by Eliezer Yudkowsky and Nate Soares without endorsing its claim that superintelligent AI inevitably brings catastrophe.
Google DeepMind did not comment on Bastian's concerns or the broader implications of his resignation. The company also did not address the open question of whether the risks of superintelligent AI outweigh its potential benefits. Bastian's resignation highlights growing unease among researchers about the pace and direction of AI development.
Source: thedecoder
