AI scientists occasionally provide warnings about the burgeoning technology’s evil side. While many people believe that artificial intelligence will eventually surpass human intellect, others argue that computers will never be smarter than humans. Ilya Sutskever, cofounder of OpenAI, offers an intriguing notion. He believes that when superintelligent robots emerge, humans may opt to merge with AI.
That seems like something out of a (very terrifying) Sci-Fi movie, doesn’t it?
Humans and AI are combining, according to the cofounder of OpenAI.
Sutskever discussed how the world must prepare for computers that might outwit humanity in an interview with MIT Technology Review. He stated that such superintelligent machines will “see things that we don’t see.”
This concept encouraged him to collaborate with Jan Leike, another OpenAI scientist, to form a team. The team is concerned with ensuring that the AI models only do what humans expect of them. The OpenAI team has also given this approach a name: superalignment. Superalignment, in other words, is “alignment applied to superintelligence.”
Sutskever told the newspaper that keeping superintelligence under control is a “unsolved problem” that few people are talking about or working on. “It is obviously critical that any superintelligence that is created does not go rogue.” Obviously.”
So, once we’ve solved the problem of rogue AI, what comes next? Sutskever believes that in the future, people may opt to integrate with AI instead. He stated that while the idea may appear “crazy” now, it will not be in the future.
“Many humans will opt to become AI. At first, only the most bold and adventurous individuals would attempt it. Perhaps others will follow. “Or not,” he added.
Sutskever also stated that he may be one of the first individuals to accept becoming part AI as he parted ways with the interviewer.
AI may pose a threat to humanity.
Sutskever isn’t the only AI specialist who is concerned about the technology becoming superintelligent or smarter than humans. Geoffrey Hinton, one of the three godfathers of AI, stated earlier this month in an interview that the growing technology might pose a threat to humans over the next 5-20 years.
Hinton stated in an interview with CBS News’ 60 Minutes that we may have “things” more intelligent than ourselves for the first time ever. He also mentioned the prospect of AI controlling humans in the future.
“They will be able to influence people, right?” “And these will be very good at convincing people since they’ll have learned from all the novels that have ever been written — all the books by Machiavelli, all the political conspiracies, all that stuff,” he said. They’ll figure it out.”