Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Facebook M – The Anti-Turing Test

medium.com

Post reply on HN
Print view View original

141–150 of 162 posts

Re: Facebook M – The Anti-Turing Test

#141

Post by azernik »

icebraining wrote: post #12
BukhariH wrote: post #8

Earlier quoted context omitted.

> "The opinion is split as to whether or not it’s a real AI, and there seems to be no way of proving its nature on way or the other." Clearly, the author didn't even do the most basic fact checking. Since, Facebook clearly told everyone that M was going to be AI that was assisted by humans. It's literally in the announcement post: https://www.facebook.com/Davemarcus/posts/10156070660595195 > "It's powered by artifici…

But that's exactly the point: is it AI trained by people, or people aided by AI tools? It's not the same.

I think there's a large space where the two are indeed the same.

Re: Facebook M – The Anti-Turing Test

#142

Post by azernik »

aeturnum wrote: post #135

What a silly conclusion. The fact that a human called his land line does not mean M (the thing in messenger) isn't an AI. At best, it proves that M has humans who work for M making phone calls. I don't have any insight or opinion about the question of how human M is, but this article seems makes a bunch of assumptions that make the whole investigation somewhat silly.

I think the distinction is a philosophical question. Are they humans who run errands for the AI, or is the AI a tool the humans use?

Perhaps better to think of them as coworkers, each specializing in their strengths.

The question is, just how much of the workload is the machine capable of handling? Because I think that's the big indicator of scalability.

Re: Facebook M – The Anti-Turing Test

#146

Post by twright0 »

kuschku wrote: post #88
andreasvc wrote: post #43

Earlier quoted context omitted.

If you want to avoid giving information about whether you are an AI or human, you simply respond "No." Besides, the CAPTCHA's that are sufficiently hard to solve for computers are already hard for humans as well.

> Can you tag for me all photos in my album that contain a kitten, but not a dog, with "kitten", and those that contain a kitten and a dog, with "pets That should keep it busy.

This approach is backwards. This is the kind of problem that is easy for a person, but not for an AI. So if M was an AI pretending to be a human, you could use this to determine that. But in this case, the suspicion is that M is a human pretending to be an AI - and they could simply decline to attempt the task, or pretend to be unable (or do a bad job deliberately), and you'd learn nothing from negative results.

Re: Facebook M – The Anti-Turing Test

#147

Post by hodwik »

hodwik wrote: post #86
jhpriestley wrote: post #79

Earlier quoted context omitted.

The translation of unrestricted human language into a restricted, "logical" language which could be understood mechanically, is another long-standing problem - going back as far as Liebniz - which is unlikely to be solved by a few gigabytes of chat logs.

This is not a problem of structural semantics, each phrase is itself a semantic atom -- think Zork. It doesn't need to see "What|WP 's|VBZ it|PRP like|IN outside|IN ?|." Just "What's it like outside?", and know that it will always respond to that with the same call.

(Can't edit.)

Warning: This is mere conjecture.

Re: Facebook M – The Anti-Turing Test

#148

Post by tonylucas »

gohrt wrote: post #94
jhpriestley wrote: post #79

Earlier quoted context omitted.

The translation of unrestricted human language into a restricted, "logical" language which could be understood mechanically, is another long-standing problem - going back as far as Liebniz - which is unlikely to be solved by a few gigabytes of chat logs.

Are you claiming that chat logs won't contribute at all? Or are you claiming that progress is irrelevant unless someone launches a 100% solution all at once?

I think the chat logs will help, but it won't just be about that data.

The data around how users interact will also be important:

Do they prefer a back and forth conversation, or do they want to say everything in one go.

Do they want to start a conversation, drop it, come back to it several hours later, or do they like completing it in one go.

How do they handle switching back and forth between different contexts, if certain requests take time, or do users not switch context.

What data are they happy to share, and what are they not.

What are the typical response times that a user considers acceptable 5 seconds, 1 minute, 5 minutes, 60 minutes? Does it vary depending on scenario.

Is there particular services or information that there is a trend towards, for example local search requests, research/information, particular types of purchase etc.

We've just spent 6 months going through a very similar process to this, which has helped drive the development of our Converse platform, which allows people to build semi or fully automated conversational messaging services, so this is fairly closely related.

(From our point of view, the NLP data we gathered was useful, but it wasn't the most important part)

Re: Facebook M – The Anti-Turing Test

#149

Post by tonylucas »

icebraining wrote: post #12
BukhariH wrote: post #8

Earlier quoted context omitted.

> "The opinion is split as to whether or not it’s a real AI, and there seems to be no way of proving its nature on way or the other." Clearly, the author didn't even do the most basic fact checking. Since, Facebook clearly told everyone that M was going to be AI that was assisted by humans. It's literally in the announcement post: https://www.facebook.com/Davemarcus/posts/10156070660595195 > "It's powered by artifici…

But that's exactly the point: is it AI trained by people, or people aided by AI tools? It's not the same.

I think that also depends on who is actually completing/fulfilling the task, in the DMV example its AI adding context to a task that can't be automated easily.

The AI can still do the requirements/information gathering and leaving the job for a person to do. For common issues (support triage, customer service issues) even this still has significant value.

Equally from a training point of view, it may simply be the person correcting/confirming the AI is right, and leaving it to get on with the process, rather than fallover to the human completely.

Re: Facebook M – The Anti-Turing Test

#150

Post by tonylucas »

TorKlingberg wrote: post #70
lhnz wrote: post #7

I wonder how many years it will be before a real AI can compete with Facebook's AI? I guess by the time that's possible Facebook's pretend AI will already have cornered the market. The public will only be able to see that you were the late entrant and that while your AI is faster it's occasionally incorrect in peculiar ways... This seems like a fairly solid plan by Facebook to crown themselves the winners of a race t…

> I guess by the time that's possible Facebook's pretend AI will already have cornered the market. And cost Facebook a lot of money. Are they planning to pay for personal assistants for everyone?

You are presuming that they will employ all of the people needed.

Why for example would they not turn this into a platform that easily added the AI benefits for external vendors and services, and just be the middleman collecting a cut in some way.

They are already starting to integrate vendors into Messenger directly after all.

Post reply on HN