In 2015 it felt like Artificial Intelligence (AI) was almost a reality. All the biggest tech companies teased the world with AI projects and AI raised important questions that had some of the most important figures of the tech industry such as Bill gates, Stephen Hawkings or Elon react. Is it going to better our future or is it going to kill us all?
To summarize how research has progressed in this field we can go back to 2012 when researchers focused their efforts on neural networks. Since then progress on AI has been fuelled by the development of those “deep learning” neural networks that learn by feeding on very large amounts of data.
Last summer we had computer dreaming with Google’s; this was one of the first times anyone, at home could interact with the “mind” of a computer (http://deepdreamgenerator.com/).
Google programmed one of its artificial neural networks, usually used for image recognition for its Reverse Image Search. So they gave the computer an image and asked it: “whatever you see on this picture, make it appear more and more?” If a mountain on a picture looks a little bit like a building, the network will make it look more like a building. This in turn will make the network recognize the building even more strongly on the next pass etc. until a highly detailed building appears out of nowhere.
Now you could ask “what is the purpose of this apart from creating trippy images?” Well it is a step, a first step in computer interpretation, creation, imagination etc. Like a human watching a cloud and seeing a fish in the sky, the computer watches mountains and sees buildings.
The world’s biggest tech companies have hired experts in AI to apply the technique to tasks such as voice or facial recognition.
“My personal challenge for 2016 is to build a simple AI to run my home and help me with my work,” said Mark Zuckerberg who plans to code a virtual system that will use voice controls and facial recognition, among other capabilities, to help out around the house. “I’ll teach it to let friends in by looking at their faces when they ring the doorbell,” he explained. “I’ll teach it to let me know if anything is going on in Max’s room that I need to check on when I’m not with her.”
AI researchers at Facebook are working on teaching basic physical concepts to a computer, like gravity concepts, forces and so on. Humans learn through experiencing those concepts when they are young. Facebook is working on its image-processing software to make it predict whether a block will fall or not, this in turn taught the software physics concepts and developed its common sense. After a series of test it got a 90% accuracy rate in predicting the event beating most of humans, according to Mark Schroepfer CTO at Facebook.
Facebook is also developing M a personal assistant similar to Siri but more powerful. The beta was launched to a limited amount of users last summer and according to some of the testers we are getting close to Samantha the AI in Spike Jonze’s “Her”. However M is not yet fully autonomous, it is learning from a team of trainers. They hope that it will end up picking up new tasks on its own.
Apple has recently acquired Emotient Inc., a startup that says it has developed a software that can read human emotions using artificial intelligence.
The startup claims on its website that they are “the leader in emotion detection and sentiment analysis based on facial expressions,” This would be of a real use for businesses in order to better understand customer’s understanding through emotion analysis.
They also acquired VocalIQ, a startup that should help Siri act more like a real person.
The VocalIQ AI has the ability to ask clarifying questions when the request from the user in unclear, like a human would.
Google, Apple, Facebook, the intention of those leaders in the tech industry is clear, to be the first possessing an advanced form of artificial intelligence.
The car industry is also about to undergo a revolution thanks to AI. Several major car manufacturers such as Toyota, Mercedes or Tesla are working on automated driving cars and on different systems assisting drivers made possible with AI. This would reduce human risk and improve security and comfort. Everyone was astonished by Mercedes’s F015 self-driving car, which proved us that autonomous cars are coming sooner than we thought.
AI has always had cynics who thought it could sign the end of our civilization. Movies often depict AI as evil, a superior intelligence that naturally wants to takeover humans (2001 space Odyssey). Thus, it is not surprising that some want to keep us away from that scenario.
Elon Musk the founder of Tesla and SpaceX fears that humans could be reduced as “pets” by the intelligent machines. Along with Steve Wozniak, co-founder of apple and Stephen Hawkings, he wrote an open letter concerning the use AI in military defence systems: “If any major military power pushes ahead with AI weapon development, a global arms race is virtually inevitable, and the endpoint of this technological trajectory is obvious: autonomous weapons will become the Kalashnikovs of tomorrow” Musk described AI as the biggest existential threat we’ve ever faced and called for regulations at an international level and Hawkings noted, “Humans, who are limited by slow biological evolution, couldn’t compete, and would be superseded.” Bill Gates also agreed with Musk and is worried that some people aren’t concerned about how it could affect our future.
All those famous scientists raise their voices because they know how dangerous the creation of an armed AI race and probably also the creation of a fully autonomous AI are. However they do not want to stop the research on AI, they just want to aim it toward a beneficial purpose and want to show the people that they have control over the risks that AI can engender.
For Vocal Views Members, the opinion on AI is divided; we conducted a survey on AI and 37% of the respondents think AI will bring a better future, 32% don’t know whether it will be bad or good and for 31% AI rimes with Apocalypse.
Current AI is not yet ready to pass the Turing test like Ava in “Ex Machina” but it surely is a matter of months until we start seeing the first sparkles of this promising but somehow scary intelligence implemented in our daily lives. AI is most probably the next groundbreaking revolution after the Internet and the next step in our evolution; lets just hope it won’t be the last…