Live data from Hacker News

Show HN: I'm a 14 year-old dev who has programmed an artificial intelligence bot

acuman.us

211–220 of 311 posts

Re: Show HN: I'm a 14 year-old dev who has programmed an artificial intelligence bot

#211

Whenever I find one of these AI chat programs, I like to start by typing in "this sentence is false", and yours gave me the most unique response (after quite some time calculating): "It seems to be, and it's actually a string of words satisfying the grammatical rules of a language." On a side note - I work in tech education, and I have to say that the software projects that motivated people of your age are creating i…

It seems the answer to "this sentence is false" also comes from Wolfram Alpha, which misunderstands and just gives the definition for "sentence":

https://www.wolframalpha.com/input/?i=this+sentence+is+false

Re: Show HN: I'm a 14 year-old dev who has programmed an artificial intelligence bot

#212
post #66

Earlier quoted context omitted.

I hate to say it but it's a (great looking, fun) chat bot, not an AI.

From a strictly technical point of view, it is at least some form of Artificial Intelligence: 1) From the perspective of the Turing Test, it biomimics human behavior and intelligence. 2) The field of artificial intelligence deals with a computerized comprehension of data. ACUMAN is heavily dependent on machine learning, natural language processing, and text classification algorithms, which facilitate understanding. I…

Edit: just realised you're the author, so I want to reiterate: this is a lovely designed app and making your own chatbot is rad. Original reply below.

Re 1. Yes, it does attempt to pass the Turing test.

Re 3. Yes, that includes interacting with a human in a similar way that a human would with another human, that's part of the Turing test.

My own experience was: it asked me how I was, I told it I had a hangover, it thought that was great and that I was 100% happy. Which wasn't a long test. Obviously your own experience might have varied.

Re: 3. Isn't machine learning just machine learning and NLP just NLP?

I personally don't feel like my app suddenly becomes 'AI' if I import nltk and start tokenising some corpus to train it with, or load opencv and tell my drone to avoid the big red thing. If the app would work out how to /train itself/ I would consider it an AI.

Eg, the drone having the objective to follow me at a safe distance for as long as possible, then working out that the big red thing is dangerous and avoid it.

Or the chat both to, say, have an objective of beating the human at the turing test for a certain amount of time, then working out how it gives itself away, then avoiding doing that.

I acknowledge that's just a feeling though, and many definitions online would qualify 'thing that loads library and needs human training' as AI.

However I can't help but think that's going to promote the Reading University 'captain cyborg' school of cranks.

Re: Show HN: I'm a 14 year-old dev who has programmed an artificial intelligence bot

#213

Earlier quoted context omitted.

So, I asked it: What's e^i(pi)+1? The response was... a little out. I'm impressed it had a crack at least. "It's 2.69740975483297316969108251300445672427640640289560... + 2.64355906408145617888606598966971861665798063847235.." Edit: Heh, asked it the square root of i, and the answer made me chuckle: The square root of you is (-1)^(1/4). I don't know why I'm expecting a bot to know complex plane mathematics and euler'…

Its response to your first question is in fact exactly right, because you didn't ask what you thought you asked. Your brackets are in the wrong place. You asked: e^i(pi)+1 which is: (pi * (e^i) ) + 1 which is: 2.69740975.. + 2.64355906... i So it was right. And sqrt(i) is sqrt(sqrt(-1)) which is (-1)^(1/4), so again, it was right (excepting that it put "you" instead of "i") So it did better than you think.

D'oh, you're quite right. In that case, top marks!

Re: Show HN: I'm a 14 year-old dev who has programmed an artificial intelligence bot

#214
post #208

Question: "Where have all the flowers gone?" Wolfram Answer: "Girls have picked them every one. (according to the 1960s folk song by Pete Seeger and Joe Hickerson) Acuman Answer: "I'm not really good with locations, so I don't know exactly where have all the flowers gone, but it seems to be Girls have picked them every one. (according to the 1960s folk song by Pete Seeger and Joe Hickerson)." Hmm...so closed source,…

Despite it obviously piggy-backing on Wolfram Alpha when it gets confused, I still think the normal responses are pretty good. "Where have all the flowers gone" was most likely added by hand to WA, just like all of the hilarious Siri easter eggs, and is no more impressive in terms of programming.

The kid's clearly good, just not as good as an entire company of professional programmers.

Re: Show HN: I'm a 14 year-old dev who has programmed an artificial intelligence bot

#217

This guy rocks. Sean Le Van: https://www.youtube.com/watch?v=0vyACnFsNm4 Bio "Sean Le Van is 13. He has given many professional concerts since the age of five years old, along with members of his family, The Le Van Family Musicians. He started as a singer in various styles and soon developed a passion for jazz improvisation and composition on the piano. He now embraces several musical genres, including jazz and class…

>> In his free time, he enjoys running and skateboarding, playing with his cocker, ...

Ermmmmm wouldn't they want to expand that into cocker spaniel...lol.

Re: Show HN: I'm a 14 year-old dev who has programmed an artificial intelligence bot

#218

Good job. Most of the response starts with negation, why is that? for example: Me: I am fine. Bot: You don't seem fine to me. Or Me: I am fine, sir. Bot: I am not fine, sir, but I am glad that you are. or Me: I like a r rahman Bot: I'm glad that you like a am rahman, though personally I don't like a am rahman that much. I'm just not a fan of something that is interesting and relevant to whats available to learn on th…

Because these kinds of test are silly and easily gamed by throwing in some misleading humor or negativity. It like what pick up artists call 'negging.' It puts you on the defense and little and throws a wrench in the works. The developer knows he can't really make something intelligence so its all rhetoric tricks like this with the occasional dipping into wikipedia or wolfram alpha.

Throwing your answer back at you, criticizing your question, or playing with humor is a fun little trick, but certainly not AI.

Re: Show HN: I'm a 14 year-old dev who has programmed an artificial intelligence bot

#220
From what I tried, it seems to be working on templates, similar to Eliza. However, in some templates it grabs data from Wolfram Alpha.

Very hard to find a non-canned response not coming straight from Alpha.

Still, a good project and obviously more APIs can be integrated. Maybe a good next step would be to mine more and dynamically put together templates. Try with simpler synonym directory and see where it goes.

Post reply on HN