A quick search on Area51 shows already two closed AI proposals, one 4 years ago, and one last year. Both were focusing on the technological aspects, and they may have come too early.

In 2015, we can read and hear about AI in many flavours, and even more and more. New hot companies mushroom, news spread, famous people like Musk, Hawking, and Gates warn us, software and hardware is open source to foster the development of AIs, etc.

The scenery looks very different from a few years ago, and even more different from the 90s, where, say, deep learning was starting.

Yet I am having some hard time framing the actual topic/focus. It might "just" be technology, but the OpenAI initiative and personal experience make me feel that there is quite a lot to Q&A on other topics as well, like the Laws of Robotics and related.

What is the most compelling topic for AI, if any?

Proposal: Artificial Intelligence

share
1  
I comment here to... call for comments. The project page says "create a discussion"; and I did not know it would be a Q&A form. The AI project is vague and needs brains to perhaps become meaningful. Right now, I can just see down votes but no substantial argument. Anyway, I try a bit everyday to find compelling reasons to proceed. At least the down votes confirm there is work to do :-) Thanks. – Eric Platon Dec 18 '15 at 5:14

AI as a subject isn't likely to attract a community you can build a site around because it is a broad term that's been applied to everything from creating synthetic beings to creating more-sophisticated algorithms that make computers better at what they do. Once you get past a few dozen questions about what is AI? and how we live with it, you're into the subfields where this belongs: philosophy, psychology, mathematics, statistics, linguistics, computer science, etc. We even created a site specifically to handle the algorithmic questions around closed proposals like AI — Data Science — and even they don't have an [artificial-intelligence] tag.

Artificial Intelligence isn't a site because it isn't a specific field that even the experts working in this space closely identify with. After a few science fiction fans and weekend philosophers answer the example questions, they'll be off to their jobs in statistics or programming… and the site will run out of steam. That's why a coherent community has never been able to form around this subject.

share
3  
There are SOOO Many proposals that this applies to. Maybe you can fit some of the concept into area51.stackexchange.com/faq – James Jenkins Dec 15 '15 at 15:28
2  
@JamesJenkins I started thinking about a broad statement/test that would better describe the types of "subjects of expertise" that Stack Exchange would consider suitable to this engine. Wikipedia-like subjects are not easy to describe, but these vote... on... other... things... proposals are becoming tiresome. – Robert Cartaino Dec 15 '15 at 15:39
3  
@RobertCartaino: And yet some... of... them... are doing fine despite your adamant opposition in creating them. – SF. Dec 15 '15 at 16:54
    
Thanks for phrasing this opinion. It is competing in my head and I believe this could help better framing the subject. What I keep in mind is that OpenAI was creating last week to answer questions on AI. Whatever it actually means, I think it will be important and raise further questions. On top of that, I am surrounded by people who misunderstand or misconceive what AI relates to. Common questions, for some reason, are "is AI part of ML ?" and "if ML is AI, so what is NLP / CV ?". I see quite some confusion here. – Eric Platon Dec 15 '15 at 22:30
    
I have to disagree with your statement regarding science fiction authors and weekend philosophers answering "all the questions". Behavior UX is a real field and one that is growing exponentially (I am actually working on a project that involves it currently). Same can be said for other disciplines that come into play now that AI and automation are becoming "mainstream economically". It's an overlapping zone, but so are so many other SE proposals. – Yisela Apr 29 '16 at 12:35
    
Sorry, if I'm late to the party(this comment is more about the newest AI proposal). "AI as a subject isn't likely to attract a community you can build a site around because it is a broad term that's been applied to everything". What about stackoverflow? That is loosely coined as a site for "questions on programming". Is that not broad? Programming, as a term; could be from help on "hello world!", to developing ios apps, to AI, to making an OS. So, I don't see anything wrong with it being "too broad". – Mythic Cocoa Jun 22 '16 at 6:18

Nowadays there is not enough demand on Artificial Intelligence topics and I believe it's going to change soon. Now we've already self-driving cars, driverless trains (already in London known as DLR, and some metro, and this will going to extend further), drones (self-flying AI robots delivering packages), so don't be surprise that in 10 years time we'll see artificially intelligent robots walking on the streets cleaning the rubbish, reporting crime, helping people, etc. Once the technology and engineering is more matured, we'll see AI robots produced in the amounts how mobiles are (iPhones/iPads) as part of everyday life and there would be strong demand for Artificial Intelligence Engineers.

Since SE is also about advertising the right jobs, I think it's going to be a good point to have such site, so it'll be according to the job position names such as Artificial Intelligence Specialist, Analyst, Engineer, Developer, Consultant and they would obviously expect such site to share their job-interest, otherwise where else they go? Asking job-related questions every day on 10+ different sites, because we didn't see this coming?

I believe it's not possible to divide Artificial Intelligence Engineer into either philosophy, psychology, mathematics, statistics, linguistics or computer science, it is actually a separate branch and I think sooner or later we'll need the site for it. Similar as DevOps has been established few years ago and now it bloomed (and it's neither sysadmin, development or operation, it's all together).

This topic may be not defined yet completely, as humanity is in early stage of that, but it will be soon defined with time by real-life challenges as part of these job related roles, as we will progress technologically further.

See: Artificial Intelligence jobs at itjobswatch

share
    
There is already strong demand for engineers and scientists working on artificial intelligence in many of the fields you mention, and many more. But expertise in making real-time systems for controlling trains doesn't make you know anything about robotics. Analyzing human behavior to detect crime has virtually nothing in common with self-driving cars (beyond CS/pattern recognition building blocks). There is never going to be demand for someone with a broad sense of all these areas without any deep expertise, and there is never going to be someone with 300 PhDs who can work in all of them. – Matthew Read Apr 18 '16 at 4:41
3  
TL;DR -- AI is not a branch, it's a tree. – Matthew Read Apr 18 '16 at 4:41
    
Population get smarter nowadays. The same you could say that you can't combine development and operations, now you have it (DevOps). People interested in AI are getting very dedicated into variety of challenges. Most of the challenges have the same base - neural networks. Having extensive knowledge about deep networks and architecture it doesn't matter whether you're working on self-driving cars or analyzing human behavior, you just know your stack (similar for bioinformatics). For broader sense, you've the technical/solution architects, so no need for 100+ PhDs. – kenorb Apr 18 '16 at 9:15

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .