in accordance with Dr. Nando de Freitas, researcher deep mind (company acquired Google in 2014) technology Artificial intelligence (AI), known as AGI (Artificial General Intelligence), “will be more incredible than a human.” Even with the constant progress in machine learning in recent years, AGI will presumably be on a different level as it will have neural networks that can be defined as “a little aware“. Will this be another step towards the robots well known from the movies Science fiction for decades?

Artificial intelligence proposes to create 40,000 types of chemical weapons in just 6 hours
Understand the dangers this technology can pose

The goal is to create even safer models, computer efficient, Fasterwith more memory It’s clear, more. Professionals hope to create an AI that can combine “family of intuition”. Clearly, despite the unprecedented results desired, there is still a lot of work ahead. those who are considered “human problems” are radically different from those previously predicted by machines, because it is not always possible to have only one correct answer.

This is the topic covered filmmakers other theme lovers. How could a machine understand the nuances of human feelings in an acceptable way? Something like that, firstly, is it possible? After all, understanding other people is not an easy task even for individuals of the same species, not to mention robots, even with advanced technologies. And before the full potential can be extracted from AGI, it will be necessary to gain knowledge from other people and from the reasoning involved..

The consequences of such innovative technology are extremely difficult to foresee these days. It is likely that future generations will have more opportunities for the possible new reality that such a powerful artificial intelligence can shape. In any case, as expected Dr. Nando de Freitas, deep mind is on the right track to realizing these ambitions. PresentlyCat”, AI capable multimodal achievementsalready being improved.

Cat”

The new AI system is known as “universal agent“. That is, according to the intention of its creators, in other words, it would have “the potential to compete with human intelligence and reason at many different levels of learning“. Among the assigned and possible roles, we have various daily tasks such as play atari games, stack of objects, image subtitles and even dialog.

(Functioning “Cats” / Credits: DeepMind / newsscientist.com)

Second Scott Reidco-author and researcherCat”:

As expected, some are optimistic with the potential of this evolution, others fear a possible dystopian futurein which humanity will be hostage to an extremely intelligent system capable of completely surpassing humanity. As shown in the image above, something at this level can be learned without the intervention of third parties, which is the foundation for development “off button” if there are any problems.

On the other hand, while some scientists are excited, others argue that the achievements made by AGInot as revolutionary as they think“; however, the counterargument to this statement is that it will be necessary, first of all, to build “knowledge basewhich can be further developed and improved.

It’s you? What is your opinion on this topic? Share your point of view in the comments box and on our social networks!

…..

Thinking about buying goods online? Discover the Save the Connected World extension for Google Chrome. It’s free and offers you price comparisons at major stores and coupons so you can always buy at the best price. Download now.

Via: canaltech.com.br, Olhadigital.com.br Source: newscientist.com