The “Neural Turing Machine” combines the way a conventional computer works with the way the human brain “works”
Developers all over the world should perhaps be concerned about their future career prospects, which today are wonderful.
The secretive British artificial intelligence startup Deep Mind, which was founded a few years ago by the young genius British neuroscientist of Greek Cypriot origin, Demis Hassambis, and was acquired by Google this January for $400 million, has announced that it is working feverishly to develop a computer that will be so smart that it can program itself.
The computer, called the “Neural Turing Machine,” combines the way a conventional computer works with the adaptable and ever-evolving way the human brain “works” (thinks). The hybrid machine, according to “New Scientist” and the British “Daily Mail,” learns as it stores new memories and then references them to perform logical operations beyond those for which it has already been programmed.
As the first experiments show, the hybrid neurocomputer manages to create its own -in principle simple- programming algorithms, which it is then able to use to classify and correlate data.
Google -with two artificial intelligence research groups at the British University of Oxford (in the respective Departments of Computer Science and Engineering), with the aim of helping computers - and electronic devices in general - better understand users.
The company is also working on developing very fast quantum “chips” modeled after the human brain, so that searches and related software can become more “intuitive” and predictive. For this reason, Google has begun a collaboration with John Martin, a professor at the University of California – Santa Barbara.
This was preceded, in 2012, by the hiring of extreme visionary engineer Ray Kurzweil as Google. Among other things, Ray Kurzweil, author of several books, predicts that, within 30 years at most, people will be able to "upload" their entire mind to a computer and thus become digitally immortal, while by 2100 they will have gotten rid of every biological part of their body, to replace it with mechanical and electronic parts.
Several scientists have expressed concerns that the explosive progress of artificial intelligence could backfire, as if computers acquire broad autonomous programming, action and control capabilities, no one can rule out the possibility that something could go wrong and escape human control. For some, artificial intelligence potentially poses a more serious threat than nuclear weapons.
Recognizing the potential risk, Google announced that it has created a special ethics committee to oversee all of the company’s research in the field of artificial intelligence, to ensure that there is no potentially dangerous misuse. This joint Google -Deep Mind committee will set a series of restrictive rules on the use of the related technology. The question is whether these restrictions will be reassuring enough for the future.
Source: protothema.gr
