Bill Gates: Fears AI, Likes Bitcoin, Loves Dogs - InformationWeek

InformationWeek is part of the Informa Tech Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them.Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

IoT
IoT
Comments
Bill Gates: Fears AI, Likes Bitcoin, Loves Dogs
Newest First  |  Oldest First  |  Threaded View
Page 1 / 2   >   >>
SaneIT
50%
50%
SaneIT,
User Rank: Ninja
2/2/2015 | 8:09:32 AM
Re: AI & Cortana
I don't know exactly what they fear but I see it as  a long slow path issue.  I don't think that tomorrow all of the robo cars are going to start running down pedestrians but I can see a scenario where we slowly give up control and direction to AI.  We start with following it's exercise advice, let it pick television programs and movies for us based on our "likes" helps us remove people from our social media feeds that bring us down and then once we rely on it we are open to more suggestions.  All of the examples I listed are available in current apps so we are already simple AI systems decision making responsibilities.  After a few years of getting used to liking the things that the AIs serve there won't be much questioning if things slowly drift away from our personal likes over to what an AI wants us to like.
Susan_Nunziata
50%
50%
Susan_Nunziata,
User Rank: Strategist
1/31/2015 | 10:44:00 PM
Re: AI & Cortana
@rradina: You nailed it. AI can only ever be as evil as the humans creating and controlling and using it. Which, as we know from events recent and long past, humans have quite a deep and abiding capcity for inflicting evil on one another.
PedroGonzales
50%
50%
PedroGonzales,
User Rank: Ninja
1/30/2015 | 10:22:21 PM
Re: Will to survive -- where does AI get it?
I think sometimes we are caugh up in the negatives of AI, but we forgot of the many benefits it could bring. I think such benefits can really impact our lives.  Live language translation, decision support tools that allow physician to better diagnose diseases to patients, etc.  Imagine a decision support system that helps us make better life choices such as buying a home, moving to a new place, changing careers, etc.  I think the next project for bill gates should be becoming a Sci fi writer.  A dystopian world where AI robots are controlling humans, sounds like an excellent plot and the software they run is  windows 98; a real nightmare.
Thomas Claburn
50%
50%
Thomas Claburn,
User Rank: Author
1/30/2015 | 5:29:36 PM
Re: Will to survive -- where does AI get it?
>It could learn the principle of survival and value of survival but the desperation to survive that living beings can take on isn't present in the system. Or is it, someday?

If we assume we'll teach our AI via machine learning, we'll just have to make some pre-applied rules immutable -- like you may not harm humans. The notion we'd let machines learn on their own -- without human approval of what's learned -- seems hugely unwise. It would essentially be inviting randomized action.
rradina
IW Pick
100%
0%
rradina,
User Rank: Ninja
1/30/2015 | 4:27:34 PM
Re: AI & Cortana
Do you think they fear the wrong influence controlling AI or AI controlling itself and eventually us?

I know this conjures images of Terminator, HAL 9000, Proteius IV (Demon Seed) and Saturn 3 (although the first three don't deserve to be desecrated by perhaps Farrah Fawcett's finest film!) but long before AI becomes capable of Pinky and the Brain evil, I see danger starting when humans grant implicit trust to AI.  For instance, a car that parallel parks by itself is a great innovation until it's hacked and maximum throttle is applied in reverse causing it to fly across a sidewalk packed with pedestrians.  What about when cars start driving themselves and a commuter logs thousands of blissful auto-drive miles and suddenly a software update causes a miscalculation at 70mph.

It's not that the AI is evil but as it gets better, will we increasingly grant it implicit trust and eventually get burned when it makes a deadly mistake?
Charlie Babcock
50%
50%
Charlie Babcock,
User Rank: Author
1/30/2015 | 4:05:49 PM
Will to survive -- where does AI get it?
It's not clear to me how AI gets out of control, even though Stephen Hawking is another worrier on this front. I understand an AI system like IBM's Watson can absorbe more information than I can and can access all of humanity's stored knowledge and try to learn from it. But I don't see where the desire to survive at any cost would come from in a computer system or robot. It could learn the principle of survival and value of survival but the desperation to survive that living beings can take on isn't present in the system. Or is it, someday?
zerox203
100%
0%
zerox203,
User Rank: Ninja
1/30/2015 | 11:56:22 AM
Re: Bill Gates: Fears AI, Likes Bitcoint, Loves Dogs
Reddit AMAs definitely have a unique personal feel to them, but Bill Gates tends to come off as a down-to-earth, well learned person in other interviews I've seen as well. He doesn't seem content to rest on his success, probably accomplishing more notable and varied goals in his lifetime than most will.  It says more about him, though, that he always endeavors to learn all about these topics (reading his responses about waste disposal was enlightening) as he's tackling them - versus just throwing money at a problem. Like others, the bit about AI stood out as a little weird to me. It's not like there's nothing to be concerned about on that front, but those of us who understand the technologies we call 'AI' today tend to agree that our future doesn't look quite that much like a scifi movie - but, it's hard to convey tone via text, and we may be reading too much into what was intended as an offhand comment.

The goals of the foundation do provide an interesting distillation of the technology we've developed in our lifetimes. What's the point of it all if we don't use it to help people in need? Not just critical needs like banking and healthcare, even very very rudimentary internet access could give kids around the world access to great educational resources and leapfrog them generations over what's available to them today. There was an interesting quote a bit farther down in the AMA - "So far we have not being able to use technology to connect people to the needs of the poorest in countries that are far away to tap into their empathy. I think this can be done but it needs some missing creativity." If I understand it right, that's meant to refer to crowdfunding. It's true and a little sad - if we can raise millions to make a new kind of phone case with a few clicks, we should be able do the same for those in need. 
Whoopty
100%
0%
Whoopty,
User Rank: Ninja
1/30/2015 | 10:37:47 AM
AI
I'm not so wary of AI that will get "out of control," as I think we're a long way off artificial organisms that can think outside of their intended function, or "break" programming, in the same sense that I can't change my genetic makeup if I choose to.

What is potentially worrying though is malicious AI created by someone for a specific task. That has the potential to be much more dangerous, but again, only as dangerous as the person that creates it. 
SaneIT
100%
0%
SaneIT,
User Rank: Ninja
1/30/2015 | 9:11:51 AM
Re: AI & Cortana
@mak63, you bring up some very good points.  The guys who are pushing AI and technology are the same guys warning us about letting it get out of control.  On one hand this sounds like they are aware that what they are doing has the potential to be misused, on the other hand it sounds like they want to be the controllers of AI and this is their entry point.  If they get out in front of the charge and declare themselves the champions of safely developed AI then they get to steer the direction of AI.  I do agree that AI has the potential to be more disruptive than taking low level jobs but at the same time I think some people would be very happy being controlled by AI.  I look at people who model their day around what their fitness tracking watch tells them is best for them, why not a smarter device that does a better job of leading?
mak63
100%
0%
mak63,
User Rank: Ninja
1/30/2015 | 2:41:09 AM
AI & Cortana
He (Gates) sides with Elon Musk and others on the idea that machine intelligence poses a risk, and doesn't understand why more people aren't concerned.

Maybe because we don't live in the same world as them or perhaps we don't watch so many movies about it. Anyway, the latest model of the Tesla P85D has a mode called "insane". Musk is not really helping his cause.

When Gates says that he's working with Microsoft on a personal agent, I thought immediately about Microsoft Agent. Remember those? This time around has to be Cortana, right?
Page 1 / 2   >   >>


2020 State of DevOps Report
2020 State of DevOps Report
Download this report today to learn more about the key tools and technologies being utilized, and how organizations deal with the cultural and process changes that DevOps brings. The report also examines the barriers organizations face, as well as the rewards from DevOps including faster application delivery, higher quality products, and quicker recovery from errors in production.
News
How to Create a Successful AI Program
Jessica Davis, Senior Editor, Enterprise Apps,  10/14/2020
News
Think Like a Chief Innovation Officer and Get Work Done
Joao-Pierre S. Ruth, Senior Writer,  10/13/2020
Slideshows
10 Trends Accelerating Edge Computing
Cynthia Harvey, Freelance Journalist, InformationWeek,  10/8/2020
Register for InformationWeek Newsletters
Video
Current Issue
[Special Report] Edge Computing: An IT Platform for the New Enterprise
Edge computing is poised to make a major splash within the next generation of corporate IT architectures. Here's what you need to know!
White Papers
Slideshows
Twitter Feed
Sponsored Live Streaming Video
Everything You've Been Told About Mobility Is Wrong
Attend this video symposium with Sean Wisdom, Global Director of Mobility Solutions, and learn about how you can harness powerful new products to mobilize your business potential.
Sponsored Video
Flash Poll