Facebook has unveiled a new piece of tech which they are calling DeepText. It is a text understanding engine, with almost human accuracy – developed by AI. Yes the computer taught itself about language and can pick up nuances such as 'bro' for 'brother' at a rate of 20,000 texts a second and in 20 languages.
Impressive? Certainly. Scary? Very definitely.
Whilst it might be useful when enquiring about a taxi for FB to pop up a local taxi suggestion, or deleting spam, the point is – where does it stop. When does it jump the thin line from useful to intrusive and also – as it gets better at understanding us and providing relevant content – perhaps we will miss stuff which we might have liked. Is our freedom not being ever more eroded whilst our 'time' is being saved?
Or is it actually going to be a FB version of Google – trawling though the trillions of posts to find stuff that is meaningful for the individual? If you are looking to buy something, FB can pull you up a friend of a friend who is selling that exact thing. Handy. It can also grab all the information and put it into a format for advertising – making it easier to sell you an advert (the easier it is the more people will do something). Hmmm, wonder why they developed that functionality?
So, that's all good isn't it?
I guess so – but I still can't help feeling that we are being pulled more into the 'Matrix' every day and that perhaps we want to make our own choices about what we see and when we see it.
What do you think? Let us know what you think – tweet us @itslobsternet