One of the most talked about questions on Quora in 2017 was – Why did Facebook shut down its own AI after its Bot invented its own language? While most of the online articles you will find today on this topic are misleading; the real reason is a far cry from what has been rumored so far.
Facebook had never shut down its Chatbot only because it started inventing its own language imposing threat that it would get out of control. In this post, we are offering our readers a more realistic report of all that you wanted to know about Facebook’s AI experiment and why it was shut down.
For those who are still clueless what this post is all about and what had gone wrong with FB, last month the Internet was abuzz with reports published everywhere about FB shutting down one of its artificial intelligence programs after it allegedly invented its own language. As the old adage goes – “media is an organized gossip“, the media wrongly publicized the story and blew it way out of proportion.
Unfurling of Facebook’s AI Controversy
While one page mentioned, “Facebook engineers panic, pull the plug on AI after bots develop their own language,” another site published, “Facebook shuts down AI after it invents its own creepy language.” And another asked, “Did we humans just create Frankenstein?” asked yet another. As if social media’s scuttlebutt was not enough, a famous British tabloid quoted the incident as “the dangers of deferring to artificial intelligence and could be lethal if the similar tech was injected into military robots.”
As per the reports, one of the Facebook’s AI research program took an unexpected turn, especially when their bots named Bob and Alice got involved in an independent conversation and were seen negotiating when something weird unfurled.
The conversation of Facebook’s AI went like this –
Bob began by saying: “I can I I everything else.”
Alice replied: “Balls have zero to me to me to me…”
Bob then reacted: “I I can I I I everything else.”
Alice responded: “Balls have a ball to me to me to me to me to me to.”
Bob continued: “I . . . . . . . . . . . . . . . . . . .”
Initially, the AI experts at Facebook couldn’t understand what this claptrap was all about. However, after a thorough investigation, they could make out that the AI bots were supposedly communicating in a language which they had invented altogether.
The Bob’s statement – “I can I I everything else” actually meant – ‘I’ll have three and you have everything else’. Subsequent to this self-ruling development in their artificial intelligence research program, Facebook shut down the entire Chatbot program.
When Elon Musk attacked Mark
While Facebook’s experts were busy figuring out what actually had gone wrong, the news had already spread like wildfire. And more fuel was added to the story, when global tech leaders like Elon Musk, Founder of Tesla started criticizing Facebook’s CEO Mark Zuckerberg openly. Elon went on to tweet this –
Immediately after Elon Musk’s Tweet, the social media at large postulated that Facebook shut down its AI the program down after realizing the vulnerability of AI.
The real reason why Facebook shut down its AI
However, some experts have recently rubbished those rumors such reports and argued that the bots of FB’s AI had never actually invented any new language. Moreover, the neural networks had simply modified the human language, with an intention of making the interaction more efficient. In their pursuit of knowledge transfer, the bots began conversing back and forth in a derived shorthand manner, which looked creepy to us.
The report now has made it clear that it was due to the error from the programmers’ end, that couldn’t train the bots to communicate, following the rules of English language. Subsequently, in a bid to learn from each other, the bots figured out an efficient way to communicate by deriving a shorthand.
Facebook did shut down its AI research program, not because the bots went rogue, but because Facebook lost interest in the research and has put this project on hold.
Happy Reading!
