Lead researcher and co-founder of the OpenAI organization Ilya Sutskever said that large neural networks could already become conscious.
it may be that today’s large neural networks are slightly conscious
— Ilya Sutskever (@ilyasut) February 9, 2022
Artificial intelligence researchers have strongly criticized Sutzkever’s statement. According to Australian scientist Toby Walsh, this topic disrupts the discussion and perhaps even the evolution of AI.
“Each time such speculative comments are aired, it takes months of effort to bring the conversation back to the more real opportunities and threats posed by AI.” wrote researcher.
Independent social technologist Jürgen Geuter also named this view is unrealistic. According to him, such a conclusion has no real basis and is “a publicity stunt to declare magical technical capabilities for a startup.”
Deep learning pioneer Jan LeKun also vehemently criticized Sutzkever’s remarks. He believes that to achieve this goal, a special macroarchitecture will be required, which is not found in any of the existing networks.
Not even for true for small values of “slightly conscious” and large values of “large neural nets”.
I think you would need a particular kind of macro-architecture that none of the current networks possess.
— Yann LeCun (@ylecun) February 12, 2022
Recall that in March 2021, one of the co-founders of OpenAI, Sam Altman, said that in the next ten years, AI will create enough wealth to pay every American $13,500 a year.
In April, a machine learning pioneer called for an end to calling everything artificial intelligence.
In May, Nobel Prize winner in economics Daniel Kahneman suggested that AI will surpass humans, but not soon.