I recently discovered a very young and fantastic french startup: Jam. To me, this kind of concept initiates the dawn of a new era. What is Jam? It's a Human-Assisted Artificial Intelligence who can help you to find a job or an apartment and even more. You only have to send your queries by sms and it, or rather she, will respond to you within commonly one hour.
The messages were translated from french.
It feels strange in the way that I started to talk with my cellphone to a partial artificial intelligence and it's very useful. It could potentially replace a friend.
I intentionally removed the links because they will be soon deprecated. Even though all 3 urls bring me on the same website, I found it revelant. It's like a friend who can always be there for you and that's why I think it's a little revolution. People could probably rely on such an intelligence very soon like they already rely on google for their researches.
Algorithms have been used for decades but all the fuss about it is due to the data we are now able to collect with our connected devices. This is how and because these informations are interpreted and computed which is controversial. In fact, they can already predict some kind of behaviors. Some police departments in the US, including Los Angeles, Miami and Nashville, are using data mining and analytics to forecast crimes. More precisely, based on social media activity, drug use, friendships, and others, they pinpoint the people most likely to be involved in future violent crimes. Hence, depending on your neighborhood, your relations and your personal informations, the police is going to be suspicious of you and treat you differently, not because you were actually involved in a crime but because you might be.
Algorithms will take more and more decisions, responsibilities and also power. One other example is the Amazon's recommendation algorithm. Based on what the user has bought in the past, items he has rated or liked and what other customers have viewed and purchased, Amazon is able to suggest her relevant items. And it works, the company reported a sales increase after their recommendation system establishment.
On the other hand, according to Kevin Slavin, algorithms can be in conflicts and lead to nonsense situations where, for instance, a book reached 23 million dollars on amazon. They are also involved with the flash crash of 6 May 2010 when the Dow Jones industrial average fell in just a few minutes, to get back 20 minutes later. Even analysts can't explain exactly how and why it happened.
What if such algo make people buy the same book, if the next best selling songs and movies are made up by a processor. Almost all areas are implied: policy, finance, economic, digital business... and with this ubiquity, we can question ourselves of the legitimity of it and who has the authority and the resources to control such processes.
Sometimes, it can be out of our control. Eventually, we'll reach a point where we will not fully understand how the decisions and actions are made. I think artificial intelligence can be very useful, and is one of the most amazing progress. People could even develop empathy for it. We have to ask where should be the limit of its control.
People will probably communicate more with non-human entities. One thing is certain, computer influence in our lives will increase in the very near future.