For the boss of DeepMind, the artificial intelligence, there are ” risks “
stock.adobe.com
AFP
Friday, 9 march 2018 18:34
UPDATE
Friday, 9 march 2018 18:37
Look at this article
LONDON | Source of multiple pledges for the research, the development of artificial intelligence also includes “risk” if it is badly used, has estimated Friday, Demis Hassabis, the boss of DeepMind, the company behind the super-computer program AlphaGo.
The possibility of an artificial intelligence that would become unmanageable, or even hostile to, is a question intimately linked to the development of this technology have to play a major role in the decades to come.
It is therefore no surprise that the topic had been discussed on Friday evening, during a session of questions-answers at the end of a screening at the University College of London of a documentary on AlphaGo, who had caused a sensation by 2016, beating Lee Se-Dol, grand master of south-Korean game of go.
“There is a whole series of philosophical questions are interesting and difficult (…) to which we will have to answer,” said the patron and founder of DeepMind, a subsidiary of Google, specializing in artificial intelligence (AI).
The AI, added Demis Hassabis, can be an ” amazing tool to accelerate the scientific research “, and constitute one of the technologies the most “profitable” in the history of humanity.
But, as with any technology, there are risks, ” -he stressed. “It will depend on how we, as a society, decide to use it,” he said.
“I believe in the power of human ingenuity to solve the technical problems (related to the IA) (…) But a more significant challenge will be to solve the ethical issues “, he insisted.
Demis Hassabis has ensured that this issue was among the priority of DeepMind, and the company had put in place concrete solutions to meet them.
The words of Mr Hassabis echo those of the famous british astrophysicist Stephen Hawking, according to which the artificial intelligence will be ” the best, or the worst thing ever occurred to humanity.”