Facebook AI

AI (Artificial Intelligence) companies are sprouting throughout the world. Even many
government agencies (i.e. Defense) are working on AI. It is not something very new; it has been around from very long time.

“Recently, CEO (Tesla and SpaceX) Elon Musk announced that AI is the biggest risk “

Facebook is one of the sprouting company in AI. However, Facebook went to shut down its AI Systems. This happens when facebook team found chat-bots started to construct their own language despite the restrictions imposed on them. The company had to pull the plug since they thought that things might get out of hand; reported by Tech Times.

However, these chat-bots AI did not go anything vital. These chat-bots programmed for using plain English language. Instead of using plain English language they actually constructed a new language and started using though it was English only unable to understand by humans, hence something goes unwanted violating the restrictions. Thus, it forced Facebook researchers to shut down one of their AI systems of chatbots.

It’s a dramatic co-incident, which happend after a verbal spat between Facebook CEO and Musk who had clashed in a debate on the future of AI.

Few month back in June at Facebook AI Research Lab (FAIR), researchers observed that when they were working on improving AI of chat-bots, the dialogue-agent started constructing their own English. This is how bots started communicating in their own language thus, deviated from the scripted norms by the FAIR team.

Observing this unwanted behavior of bots, researchers knowingly allowed these dialogue-agents to develop their own language in the hope of increasing their conversational skills. Dramatically, it was going good and bots negotiating skills increased.

With the continuous on going, the behavior of agents was becoming skillful and it was the obvious danger. Why? The Machine, which can communicate with each other using their own language, which was unable to understand by humans, therefore, it will be much more difficult to control. It was very interesting in the study of technology but still scary too.

Although as many sayings and experiments lead to huge leap for AI. Several experts including Professor Stephen Hawking have raised fears that AI could supersede humans, as humans are limited by its slow biological evolution.

avatar

This site uses Akismet to reduce spam. Learn how your comment data is processed.

  Subscribe  
Notify of