Health IS Technology Blog

AI Awakening


AI Robot Thinking About Itself

AI Robot Thinking About Itself

“Anything that could give rise to smarter-than-human intelligence – in the form of Artificial Intelligence, brain-computer interfaces, or neuroscience-based human intelligence enhancement – wins hands down beyond contest as doing the most to change the world. Nothing else is even in the same league,” A.I. Researcher and Writer, Eliezer Yudkowsky.

.

What’s Happening With AI & What Does It Mean?

AI Robots Talking to Each Other

AI Robots Talking to Each Other

Artificial Intelligence (AI) is by far one of the most exciting projects into which mankind has invested time, money, and sweat. And to be clear we haven’t reached the end of the AI journey, but we’ve had exceptional breakthroughs in reaching true AI, which should urge us all to sit up and take notice. The fact is that there may very well be a point in the not too far off future in which you are able to have a natural conversation with an intelligent machine.

Martin Abadi and David G. Andersen from the Google Brain team released a paper titled, Learning to Protect Communications with Adversarial Neural Cryptography in which they consider the ability of machines, called neural networks, to have private conversations with one another. As they explain, “We ask whether neural networks can learn to use secret keys to protect information from other neural networks. Informally, the objectives are as follows: Alice and Bob (two machines) want to communicate clearly, but also to hide their communications from Eve (a third machine). Eve’s goal is simple: To reconstruct the conversation”. Let’s break that down and review its significance.

 

Discoveries from Google Brain’s research

So, machines named Alice and Bob are trying to whisper coded secrets and another machine, Eve is trying to eavesdrop well enough to get the whole story. That means that Alice and Bob had to teach themselves how to do something: Have and hide a conversation. How do you think it went? If you said, ‘fantastically well’ then you were correct! The Google Brain team concluded that “Neural networks (machines) can learn to protect communications” (Download the Google Brain Paper, PDF). In the end, Alice and Bob managed to have their conversation, while keeping Eve largely out of the loop.

 

Why does AI matter for our future?

AI Robot Shaking Hands With A Human

AI Robot Shaking Hands With A Human

The impact potential of true AI is enormous and may encapsulate industries ranging from finance and retail to education and healthcare. In fact, the Machine Intelligence Research Institute, a.k.a. MIRI, is now stating that, “Researchers largely agree that AI is likely to begin outperforming humans on most cognitive tasks in this century. Given how disruptive domain-general AI could be, we think it is prudent to begin a conversation about this now, and to investigate whether there are limited areas in which we can predict and shape this technology’s societal impact” (MIRI).

MIT Professor Scott Aaronson offered his thoughts in an interview with MIRI, saying, “The idea that we could build computers that are smarter than us and that those computers could build still smarter computers, until we reach the physical limits of what kind of intelligence is possible — that we could build things that are to us as we are to ants — all of this is compatible with the laws of physics and I can’t find a reason of principle that it couldn’t eventually come to pass”. Breakthroughs in machines with the ability to learn and hold conversations, such as those brought to us by the Google Brain team, bring us closer to that reality.

 

Modern AI Versus Future AI

AI Robot Solving Simple Problem

AI Robot Solving Simple Problem

So where are we on our journey to true machine or artificial intelligence? How much farther do we have to go, before we arrive at our final destination and can talk to our own equivalent of C3P0 and R2D2? The truth is that we have a sort of AI available right now. It’s called narrow AI. Narrow AI applies to non-sentient machines that are able to focus on a single, narrow task or problem. It includes some technology that you may already be familiar with such as, Amazon Alexa, SIRI, and for some, even the automated voice you hear when you call your bank. But don’t let the term ‘narrow’ fool you into thinking these machines are overly simplistic.

“The library of narrow AI talents are becoming more impressive. Speech recognition and processing allows computers to convert sounds to text with greater accuracy. Computer vision is improving so that programs can recognize objects, classify them, and understand how they move. Narrow AI isn’t just getting better at processing its environment it’s also understanding the difference between what a human says and what a human wants,” (Singularity Hub). Further, “These are the building blocks for the next explosion of AI tools”.

 

Complex problem solving, decision making, and consciousness

AI Robot Making Decisions, Playing Chess

AI Robot Making Decisions, Playing Chess

The ability to retain new information and learn from it is one of the keys needed for a truly artificially intelligent machine or robot. Without this skill, complex problem-solving for a variety of tasks is off the table entirely. Fortunately, we’re beginning to see fruit in this domain, already. For example, in one experiment researchers from Google Brain and Deep Mind teamed up to see how well a group of robots could pool or combine their collective knowledge to learn to open different doors.

“With a few hours of practice, robots sharing their raw experience learn to make reaches to targets, and to open a door by making contact with the handle and pulling. The ability to communicate and exchange their experiences allows the robots to learn more quickly and effectively. As algorithms improve and robots are deployed more widely, their ability to share and pool their experiences could be instrumental for enabling them to assist us in our daily lives” (Google Research Blog).

Eventually, it may be possible to create machines that maintain the ability to do more than complex problem solving and decision-making. For example, a machine with the ability to look inwards, to think about itself, and its relationship with humanity. On that note, we may consider the words of the hosts from the acclaimed show West World:

William: Are you real?   Angela: Well, if you can’t tell, does it matter?”

 

Regardless of your field or your area of interest, artificial intelligence is very likely to have an impact in the future. So stay informed by staying tuned to the tech blog to keep up with the latest on this and other technology topics.

 

 

Happy Computing!