Overheard in the AI World

Original

Artificial Intelligence already has a huge impact on our lives daily, be it at home or at work, and as this technology continues to improve it's only the natural progression for AI to become even more commonly used across all industries. Tractica predicts that annual Global AI enterprise software revenue will grow from $644 million in 2016 to nearly $39 billion by 2025. With these advancing technologies, businesses will become more intuitive but how will AI impact all industries? 

At each RE•WORK summit, we’re surrounded by global leaders and experts in AI, presenting the opportunity for us to not only have some interesting discussions but to overhear some fascinating exchanges. Our attendees and speakers alike discuss their upcoming and current work, as well as speculating about the future of AI. Knowing that you wouldn’t want to miss out, we’ve caught some of the conversations we’ve heard.

Here’s what our attendees had to say:

What's the difference between Artificial Intelligence and Artificial Emotional Intelligence? How can a machine be 'emotional'?

“Emotions are like little bars from 0-100% that you feel. All of them are little containers that get filled up, and then influence your behavior. Brains are very sensitive systems, which means even a small influence can change a persons behaviour. So what AI needs to do, is have a theory of mind which allows it to tell ahead of time "touchy" subjects. This is emotional intelligence - manipulating entire minds, extremely complex systems, by doing/saying the right thing at the right time. A lot of non-emotional intelligence uses logic because non-mind systems tend to act like a machine. A rock in the air will always fall the same way, a tool like an axe will always work the same way. So you can manipulate this machine, fully understanding it. A mind can do anything at any point, and it can surprise you. That is why it seems like there is free will because most of the mind is hidden away. Emotional intelligence is all about getting key information about a mind, and using that to make a simplified model.”

A lot of concern around AI appears to focus on human-level or “superintelligent” AI. Do you think that is a realistic prospect in the near future?

“From what I've heard, it's very possible. I'm hoping we have something that can at least pass a proper, harsh, skeptical Turing test by 2100.

Consider this: AI is very good at specialised tasks. If a sufficiently advanced AI was created solely to create another AI, it could create one better than itself. And this could continue almost infinitely.

Also this: what we consider human-level AI is based on our own consciousness, which has fundamental differences with AI. For starters, our minds run on hardware based on chemistry and electricity while computers just have electricity. What I'm trying to get at is, once this fundamentally different AI is as good as a human at being human, it's probably so advanced to achieve this feat that it's by far better than a human at everything else.”

 As AI gets increasingly smart, who should be responsible for the decisions it makes - the machine or the developer?

“I think it's up to the developer to properly test the system before deploying it. AI's are not inherently aware of ethics and are easily duplicated. It would be completely impractical to prosecute something like that.”

“You can't lock it up, it doesn't have any property to seize or money to pay fines. It's like holding your toaster responsible for burning your toast. It might feel satisfying, but the toaster isn't going to change its ways because you cursed at it.

Until you can get Captain Picard to successfully argue the citizenship of an AI and thus allow it to own property instead of being property, the whole idea of blame is meaningless.”

“When it can 100% pass the Turin test under any conversation scenario, I would say it should be held responsible for itself. Until then, the responsibility is on the developer.”


 Which industries do you think are going to be most transformed by AI in the next 5 years?

“I see law and medicine. If we get decent NLP understanding coupled with the image recognition we have today, medical diagnosis will be doable by a smartphone app.”

“With any luck, we won't reduce the number of doctors but increase the time they can spend with patients who actually need them. AI can handle all the little stuff ("you have food poisoning, here's a note so your employer pays sick leave", "Your child has chicken pox", etc.) and real doctors can spend their time handling severely injured, severely/terminally ill, or those suffering from mental illness. Doctors are under so much pressure to handle large volumes of patients, many of whom don't want/need to see a doctor really. It would be great to have those people handled by AI.”

 What areas of healthcare do you think will most benefit from AI in the short term?

 “They're already studies which show that AI is better and more accurate at diagnosis. Imagine how this would affect the industry. Normal doctors could get outsourced altogether. The AI could diagnose you and write you a prescription all by itself. Or send you to a specialist to get ‘x’ operation. Doctor visits would go down in costs. With an AI capable of diagnosing multiple patients at once. Just go and sit in the room that can x-ray all of you, monitor your vitals, breathing, blood pressure all from sight. It can go through family history or even genetics perfectly. Imagine being able to take a blood test and have a perfect diagnosis in 5 minutes or less.”

“The DNA testing, if done from birth for everybody has a lot of potential for catching and treating conditions that won't show up until your 40 or 50 years old. The UK NHS is not about making money, so 'treating' what might be a very costly condition later in life by a simple change in lifestyle as a child or young adult is not going to be a bad thing.”


If you want to be a part of our future discussions make sure to attend our future summits where you can learn more about applying AI, ML and DL in your business. At the RE•WORK Applied AI Summit in Houston this November 29 - 30, we will explore real-life AI applications, case studies, business insights & results from leading companies leveraging AI to solve problems in enterprise including experts from Tractica, Reddit, Netflix and more.

Original

Big Data Machine Learning Deep Learning Healthcare AI Government Transformation


0 Comments

    As Featured In

    Original
    Original
    Original
    Original
    Original
    Original
    Original
    Original
    Original
    Original
    Original
    Original
    Original
    Original

    Partners & Attendees

    Intel.001
    Nvidia.001
    Acc1.001
    Ibm watson health 3.001
    Facebook.001
    Rbc research.001
    Mit tech review.001
    Graphcoreai.001
    Maluuba 2017.001
    Twentybn.001
    Forbes.001
    Kd nuggets.001
    This website uses cookies to ensure you get the best experience. Learn more