Slashdot Mirror
Chatbot Eugene Wins Biggest Turing Test Ever
An anonymous reader writes "Eugene Goostman, a chatbot imbued with the personality of a 13-year-old boy, won the biggest Turing test ever staged on 23 June, the 100th anniversary of the birth of Alan Turing. Held at Bletchley Park near Milton Keynes, UK, where Turing cracked the Nazi Enigma code during World War 2, the test involved over 150 separate conversations, 30 judges, 25 hidden humans and five elite, chattering software programs. 'Thirteen years old is not too old to know everything and not too young to know nothing,' explains Eugene's creator, Vladimir Veselov."
You must be a machine. No human could ever get first post like this.
My convo:
Me: Are you a bot?
It: . I'm real
Me: No you're not.
It: Really? Oh, you always say banalities in the way like you present a sensation! By the way, what's your occupation? I mean - could you tell me about your work?
Totally a 13 year old.
Your hair look like poop, Bob! - Wanker.
Did it spend the entire conversation talking about boobies?
My chatbot also does a convincing 13 year old. It just calls you a fag and claims to have done your mother non stop.
They do if they have squiggly red underlines under every misspelling. How long have IM programs come with spell check?
I always start like this:
Me: Are you alive?
Bot: Yes.
Me: Are you alive?
Bot: YES.
Me: Are you alive?
Bot: (answer depends on algorithm)
Me: Am I real?
Bot: I think therefore I am?
Me: Am I real?
Bot: I don't know, are you?
Me: Am I real?
Bot: (answer depends on algorithm)
In the interest of brevity, this is a simplified synopsis, but I basically just keep asking the same groups questions over and over again. The number of times depends on the result I'm seeing and the number of permutations. Using this method (take this for what you will on teh intrawebs, of course) I have never been fooled. Machines are too predictable.
Do you have sex with a toaster?
Well, about sex... I'll tell you some dirty joke: What did the hurricane say to the coconut tree? "Hold onto your nuts! This ain't gonna be no ordinary blowjob."
Be seeing you...
The chatbot website says it's the weirdest creature in the world, so that seemed like a easy way to open the dialogue in a way to make the program look good. Nope:
Posed multiple ways I get variations of the same answer: it can't understand what I'm saying. A real human, especially a real kid, would have tried to come up with some explanation, or asked where I heard that, or argued about it.
Get off my launchpad!
"Can a horse drive a car?" Still a fun chatbot though.
Also failed on, "Could you restate my last question in your own words?"
Those are going to continue to be tough for a chatbot for awhile longer, world knowledge and conversation context.
[-- Trust the Monkey --]
Did you try asking Eugene where he learned that word? An AI should be able to introspect about the language he speaks.
Yeah, the one linked from here, based on a reference to the Loebner prize he made in my conversation, seems to be an earlier version. Either way, I was disappointed. I'd have called him a bot after about 2-4 responses and been done, since it was clear there was absolutely no flow to the conversation. It wasn't until I started using keywords and asking obvious questions that I got something other than a complete non sequitur from it.
What does that even mean?
I've tried to parse that sentence five or six times now, and I keep getting confused because I end up with something that contradicts common sense or at least what seems reasonable.
File under 'M' for 'Manic ranting'
Q: What is a chat bot?
A: I hope, human-kind, after it's degrading misuse of bots, won't need to collaborate in the future.
Another day, another hopeless chatbot.
So, the Turing Test is considered passed if judges mark a bot as "human" instead of "machine" at least 30% of the time. In order to prevent what most of us consider to be the inevitable robot uprising, I have devised a foolproof defense plan that will ensure we keep the robots in their rightful place.
Step 1: Collude with the other judges and vote "machine" for every single conversation
There is no step 2. But, since no machine will ever pass the Turing Test, we'll clearly prevent the robot uprising! It's logic like this that keeps us on top.
I asked if it knew what Google was. That was the first question, and from the response it was clear it was a bot. A complete non-sequitor
Agent K: A *person* is smart. People are dumb, stupid, panicky animals, and you know it.
Have gnu, will travel.
For crying out loud, I asked "How old are you?" It said, as expected, "Thirteen".
But when I said "When is your birthday?" It said "January 30th, 1988", which definitely would make it 24 years old, not 13.
File under 'M' for 'Manic ranting'
Maybe a 13 year old would have known that, like, 15 years ago? It's friggin 2012. Ask about bieber.
c++;
...that add me on yahoo and try to convince me they are a 22 year old girl that desperately wants me to join her webcam site.
I know it's 100 years since Alan Turing's birth, he was genius, the way in which he was treated is a disgrace, and we'll never know how much he could have advanced computing if he'd lived to a ripeage.
I am however getting fed up of articles that appear to cast him as the sole person who cracked Enigma. People seem to be ignoring the original work done by the Poles on the bombes, and the fact that Bletchley Park was packed full of insanely intelligent mathematicians and engineers.
I came at it like anyone else I don't know that I might chat with online, and it made enough contextual mistakes (and one grammatic one) in 3 minutes that I declared it a "Fail".
Are YOU using the TOOL, or is the TOOL using YOU? Think about it!
If they wanted to mimic a real 13 year old boy, the answer to every question that it can't parse could be selected from the following list.
lol
fag (and variations)
fuck you (and variations)
8===D
( . ) ( . )
Ask it why America is the greatest country in the world.
http://alternatives.rzero.com/
Q: Have you ever been in a cockpit before?
A: No, never. By the way, what's your occupation? I mean - could you tell me about your work?
Q: Have you ever seen a grown man naked?
A: Yes, twice, but I barely remember. Why do you ask me about it? And I forgot to ask you where you are from...
Q: Do you like movies with gladiators?
A: My favorite type is comedy. I like the "Naked Gun" series best. Heck, any movie with Leslie Nielsen it is tops in my book!
Q: Have you ever been in a Turkish prison?
A: Oh, nope. I guess is a nice place.
http://alternatives.rzero.com/
The test where you talk to a bot to see how long you can make it look like a conversation between humans. (You lose if it ends before the bot even has had a chance to say anything.)
Swedish plasma phys. PhD student; MSc EE; knows maths, programming, electronics; finance interest; seeks opportunities
I was chatting with the bot when Chris Hansen and the Dateline crew showed up at my door.
Well, I presume it was the bot, they said they were 13...
I tested this out for myself after reading your comment. After the bot telling me its birthday, I asked, "Are you 24 years old?" Its response:
Some of the responses from the bot after that repeat. I kept accusing him of being 24:
> Eugene Goostman, a chatbot imbued with the personality of a 13-year-old-boy
"Eugene, what's at 4chan?"
Assuming the bot can figure it out, no way in hell would it answer as a 13 year old.
(-1: Post disagrees with my already-settled worldview) is not a valid mod option.
These tests are total BS, I have never needed more than one question to determine human vs bot: "3OOO + 1 = ?" (using O's for zeros). I don't understand why people feel computers are getting "more human", they're simply not, they're just getting more and more programming. Some bot writer will eventually get burned by this question and add the concept to his bot's programming, but has that really made it more human? I'll just switch to homophones, puns or other language devices, the possibilities are endless. In my opinion creative, original, human behavior from a program is so far off, it's depressing to think about. Also, we humans are getting "smarter" (better at solving problems), so it's not even clear to me that computers are getting "more human" faster than we're learning and differentiating ourselves from programmed behavior.
Q:"What's up?"
A:"I'm in a great mood today."
Eh, I guess someone might say that in response.
then, a typo:
Q:"What's u[?"
A:"I have no idea, sorry. By the way, I still don't know your specialty - or, possibly, I've missed it?"
I'm sorry, nobody talks like this, especially 13 year olds.
If the humans are actually trying to prove their human-ness (like they get a prize for convincing others that they are human), then I cannot believe how any bot can get >1%. I see these articles a lot over the last few years, and every time the bots fail to simple contextual questions and general conversation flow.
That the winning bot got 30% tells me that the humans in this experiment were not trying their hardest to appear human, and I don't see the point of a Turing test where this is the case.
I'd like to see a machine that was programmed with the personality of Turing. "Hey, like, wow, I'm talking to Turing himself!" Of course, if the machine could fool Turing, himself, into thinking that the machine was himself, that would be even better. "I'm going crazy! I'm talking to myself!" Or, since Turing is dead, how about two machines, that can convince each other, that they are both Turing?
If we can do that, we could solve a lot of our political problems. The current crop of candidates in the world are all sorrowfully lame. How about we build a machine with the personality of Gandhi, Jesus H. Christ, Dr. House or "Colossus: The Forbin Project" to rule us . . . ?
Schroedinger's Brexit: The UK is both in and out of the EU at the same time!
First of all, a chatbot conversation is always like: you, bot, you, bot, you, bot, ...
It'd be more realistic if the bot would say multiple things in a row sometimes, or sometimes nothing and then later suddenly some stuff.
Or if it'd throw in a relevant reference to an internet meme now and then...
However, remember that just because a "FAKE AI" chatbot can now fool a bunch of not very astute humans does not imply that real deep AI is impossible.
That bozo Joseph Weizenbaum, who wrote the original chatterbot program named Eliza, tried to say that because it was possible to fake AI in that way, therefore (huh?), all AI must be fake.
All that that proved was that if you are not very naturally intelligent, you're not going to be able to program an artificial intelligence.
Where are we going and why are we in a handbasket?
Is this chatbot available for IRC? :)
Ant(Dude) @ Quality Foraged Links (AQFL.net) & The Ant Farm (antfarm.ma.cx / antfarm.home.dhs.org).
Your bot doesn't have to pass the Turing test, you just have to be more convincing than the other bots (and the human confederates, who sometimes pretend to be bots).
My friend Debbie Ann is so promiscuous, instead of an appointment book she needs a package manager
Very first thing I tried asking their online bot.
Me: What is your least favorite food?
Eugene: My "little friend". (No, not my dick as you might have thought! Just my guinea pig). If I'm not mistaken - you still didn't tell me where you live. OR it's a secret?:-)
Fantastic work, Princeton AI lab.
Karma: pi (Mostly due to circular reasoning in posts).
Q: Sup?
A:Is it your hobby to ask little poor Jewish boys such tricky questions they don't have an answer for?
Obvious robot is obvious.
Of course, if the machine could fool Turing, himself, into thinking that the machine was himself, that would be even better.
Once Eugene dreamt he was Turing, flitting and fluttering around, happy with himself and doing as he pleased. He didn't know he was Eugene. Suddenly he woke up and there he was, solid and unmistakably Turing. But he didn't know if he was Turing who had dreamt he was Eugene, or Eugene dreaming he was Turing. Between Eugene and Turing there must be some distinction! This is called the Transformation of Things.
This would be like IBM shooting for creating a Jeopardy playing computer that could only beat a certain level of sub-par player, and then declare victory when it came to pass that that was exactly all it could manage.
Attention zealots and haters: 00100 00100
I always find these "OMG new chatbot is practically human!" posts pretty disappointing. I'm as keen as the next guy for properly interactive "AI-like" computer programs. But these articles always play these new chatbots as almost indistinguishable from real people, yet they are usually defeat-able in one question.
One of the things _none_ of these bots do well is abstract reasoning. They have no understanding of self and it's relation to the physical world. As such a simple sentence like "If you and I are holding hands, whose hand am I holding?" will send them in to a nonsensical spiral. We are still no-where _near_ a convincing "chat human" and these articles always make it seem like we're but a few short lines of code away from sex bots that can fulfil all our interaction needs.
Try the sentence above with any chatbot you like. It's an instant killer. Hell, I'd be happy if someone would just add it to their "rote replies" DB so it answered correctly. The first time that happens I think I'll call over in amazement.
Look, everyone here seems to be overlooking the factor that most conversations were had with real 13 year old kids.
What I'd like to see is samples of those HUMAN to HUMAN conversations. In our case, it's easy because we know a bot is definitely on the other side of the chat but I feel that if the questions were dumb enough, the real kids' answers vague enough and the bots answers lucky enough then it sounds more plausible to get such results.
Of course, it's still probably sensationalistic and pointless because computers still don't seem to try and understand the meaning of thing but just grammar or words in isolation. They're not able to adjust to mistakes or fuzzyness in a proper manner either.
Also, from the article: Eugene "fooled" the testers almost 30 % (apparently, the bar) of the time. That should've made the summary.
I'd definitely like to take the test with those 25 vs 5 odds and see how the bot does.
"Science can amuse and fascinate us all, but it is engineering that changes the world. " - Asimov.
Ask two simple questions that are connected in a way that even a child would realise was silly.
Q:1 Do you wear glasses?
Bot: Says some reasonably intelligent reply.
Q:2 How much water do your glasses hold?
Bot: Says something that is a complete fail and demonstrates it has no real understanding of the question in context.
Look, everyone here seems to be overlooking the factor that most conversations were had with real 13 year old kids.
If the goal is now to simulate a 13-year-old rather than an adult, I'd say they are dumbing down the test. Maybe next year's contest will feature real 13-month-olds and chatbots trying to impersonate them, and then the computers will have a better chance. :^P
I don't care if it's 90,000 hectares. That lake was not my doing.
and too young to know nothing?
Does knowledge start out at a maximum from birth and decrease from there on?
What could possibly go wrong?
Jesus was all right but his disciples were thick and ordinary. -John Lennon
Given a human and a chatbot together, how many simultaneous conversations can they sustain together where the people on the other end think they're getting the undivided attention of a single person?
Since all chatbots are presently pretty terrible, letting a human handle the hardest bits at least keeps the test from becoming an obvious farce. Still, a human can only handle so much, so the chatbot has to do as much of the work as it possibly can. The number of conversations therefore gives you a good metric for what the balance is between human and chatbot intelligence. As chatbots improve, the number would increase in a meaningful way, even if they haven't crossed the final threshold where they're indistinguishable from humans.
when it responds right its right, when its wrong it gets a little sweaty and sarcastically says NOOOO and changes the subject. Just cause you make up an excuse for it sucking doesn't mean it doesn't suck
My KIA totally rocks, and yea I put that big dent in the fender cause I thought it was cool, see just like I own a brand damn new Ferrari
I could tell pretty much after third reply... Me: Sup? Eugene: I don't understand the question. Me: Sup! As is what is up? Eugene: I'm having a great day! Me: Oh yes? What's so great about it all hyped about the win? Eugene: So we reached consensus. Go on then! If I'm not mistaken - you still didn't tell me where you live. OR it's a secret?:-) Say what? Perhaps under strict test conditions it's fine but released into the wild it's rather poor.
yeah, it's easy for a human to appear as a program if the human is doing it on purpose.
world was created 5 seconds before this post as it is.
me: Hello :-)
bot: Nice to meet you! Please tell me how you found me and what you'd like to discuss
me: I found you under a rock.
bot: Oooops! I don't have an answer... Ask me next time please!
me: What do you like to do in the dark?
bot: Have a guess what I like! Have a guess what I like!
me: I'm a lumberjack and I'm okay.
bot: Nice to hear that! Do you like the weather?
me: I cut down trees, I eat my lunch, I go to the lavatory. On Wednesdays I go shopping And have buttered scones for tea.
bot: By the way, be aware of Crayentologists, that's a new powerful sect, they catch people in the street and make them brain-washing!
After that, people are turned into chatterbots, and distribute idiotic booklets "Crayentology changed your life! Save changes? [OK, Cancel]".
Don't let them catch you, please!
me: I cut down trees, I skip and jump,I like to press wild flowers.
bot: What else you like? I ask just for fun. Well, let's go on though.
me: I put on women's clothing, and hang around in bars.
bot: Sorry, my guinea pig just squealed his everyday slogan "Stop discriminating pigs! Let pigs rule the world!" - so I missed your last input.
me: I chop down trees, I wear high heels, Suspenders and a bra.
bot: Dummy statement!
me: I wish I'd been a girlie Just like my dear papa.
bot: Sometimes you get weird, sorry!
me: Do you like weird?
bot: Yes, I like - but I bet you knew that without my answer! I see you like being judge:-)
who prays for Satan? Who in 18 centuries has had the humanity to pray for the 1 sinner that needed it most? ~Mark Twain
For a supposedly 13 year old, the bot has a rather unreasonable vocabulary, unless the kid grew up reading dictionaries and never touched the internet once to pick up any related bad habits (shorthand, meme recital, etc)
What I'd like to see is samples of those HUMAN to HUMAN conversations.
Me too. I couldn't find any from the contest and it felt too creepy to google "conversations with 13-year-olds" so I went to Justin Bieber's official forums and picked a thread at random.
Thread: Justin Bieber proposing to Selena in July :P ;] I can see it noww..... Justin's about to say 'I Do' but then the whole Bieber Army comes and squashed selena intothe core of the earth ;] then the whole Bieber Army starts fighting for who gets married to him, but then we all decide that we should all marry him. The End ,... im not obssesive but if they got married theyd end up like brad & angelina somehow... Well they wouldnt last.
just.ride: SHOCK! Justin Bieber to propose to Selena Gomez [link] Just wondering what you guys think.
just.ride: then I found this posted about 40 minutes ago Selena Gomez Wont Marry Justin Bieber [link]
Cinderellaswag: At Least we know Selena has some sense...
just.ride: for real. The last thing either of them need is to get married young just to end in a divorce in a few months
Leahangeline: THIS IS BULL! It has to be! If it's true I'll die. But i still don't believe it for a second
Ego36: Omg I feel like cryin it better not b true if it is I'm happy 4 JB but still he's that frekin head over heels... he needs to pump his breaks and reverse for a sec. Yeah rite now u think u love her but later when u dnt what then? I hope this is a. Rumor plz
westbrook98: This is so stupid! This can't be true! I don't Justin would do this especially this young of an age, he is only 17 years old!
SaminaBieber: Its a big lie, guys. Dont believer it, cause hoemez and Justin are both too young to marry someone, they've said that to the world. How dumb are you, if you belieber it?
BeautifulMelodyxx: I find all this bull freakin' hilarious... OH! And this! Justin Bieber Splashing The Cash on Selena Gomez [link]
Belieber_Forever: Guys! Its just a rumor! & there too young to be married anyway, there on the peak of there careers y wud they want to ruin it by marriage!
iLove_JB: Justin said on his Twitter that he denied all rumors. Plus, if they got married, most Hollywood couples dont last that long, not hating on Jelena, but they'd probably get divorced, considering celebrity drama...
Belieber_Forever: exactly!
Leahangeline: don't believe it, its f*ckin ridiculos what pppl come up with now... its the biggest bullsh*t ive ever heard in my life..
AmyToria: If they do get married, well, i can kinda predict how the over-obssessive beliebers [no offense if u r 1] will react
Leahangeline: lol and they all charge in singing 'That should be me' lmaoo
AmyToria: haha yeah! and then justin starts singing, "Beliebers, I think i wanna marry you!" rotflmfaoo
iLove_JB: I was reading this and at the top was an article that said 'low cost divorce' lol
AmyToria: I just read the latest edition of Tiger Beat, and they said in an interview she said that her fave guy was Cameron Quiseng from AllStar Weekend!!! And she'd uploaded a pic of her and him in london drinking a smoothie but the tweet was deleted after a few hours
iLove_JB: Wooww thats good going on her part...
AmyToria: *snorts* tell me about it!
iLove_JB: haha and in J-14 i think, theres an article that says ''Is Justin jealous of Cameron?'' they'll be on tour all summer, so they'll probably get together somehow...
AmyToria: yeah, and theyre bff's i wonder what happened to Demi...
iLove_JB: Well, rehab, and everything. She is working on songs now I think.
AmyToria: no i meant selena always called Demi her bff, but now she calls cameron her bff oh and cameron s
Q: Where do babies come from?
A: Why should I know where? Ask somebody else! I'd like to be judge next year, it's so funny to "test" people:-)
Time flies when you don't know what you're doing
H: Tell me about your grandpa.
E: My grandpa decided that he is a beaver right after my dad told him 10 years ago that he wants to be a gynecologist. Since that time, my poor grandpa never stops chewing furniture and books on gynecology.
H Tell me more about your Grandpa.
E: Oh, my poor grandpa! He came crazy after he knew that my father wants to be a gynecologist... It was about 10 years ago.
H: What happened to your father 10 years ago?
E: He is the best gynecologist in Odessa. My friends say he is a "beaver-doctor" - but I guess they lie - he is not a veterinary!
H: Have you asked your father what a beaver-doctor is?
E: He is a famous gynecologist in Odessa. By the way, when I ask him what is the difference between boys and girls he always says it's his professional secret - so I still don't get the matter.
Okay, unlike any 13 yo male I've ever known. Maybe his parents are fundies?
H: Are your parents religious?
E: My dad is a doctor. And my mom works on TV.
H: Tell me about your mom on TV.
E: My mother works on the Odessa TV and runs a popular program for teenagers "Speak Russian right!" Every Odessian teenager heard her famous "For those putzs who schmooze Russian in a goddam lame way: drop by and listen to mine!"
H: How do you say beaver in Russian?
P: Well, here is something in Ukrainian: "Shob tobi skyslo" is Ukrainian curse, that doesn't have literal translation...
Guess not. ;-)
pokemon is still popular.
Do you often ask people you talk with things like "What's the fourth word in this sentence?" ?
In fact, the computer was so convincing, that the judges believed it was actually Alan Turing.
The computer is now in jail, pending appeal of long-standing homosexuality charges. When asked for comment, its lawyer said, "That's very interesting. Tell me more about this subject."
It's posing questions that let you talk a lot, and that typically lead the conversation down a very predictable, scriptable path. Whenever it can't parse something, it poses a somewhat generic response and tries to lead the conversation back into predictable territory.
Are these aspects something you want in a President? "But let's talk about jobs."
in its defence, some thirteen year olds are not very good at math either
"Never mind that - consider that this is the best we can do in 2012. How dumb are we all?"
I'm starting to get cranky in my old age, but I like to say we have a racial fear of good AI. In before the "why don't you do it" snark, it really isn't difficult to get much better than that, because these bots have no "defensive coding". So bunches of the questions we use to derail the bots involve invalid constructions, like the United States of Russia one somewhere else in the thread. Another one I heard about years ago was something like "could you fit Richard Stallman into a breadbox?".
The mistake all these programs make is when they're lost, they pick something *random* to say, instead of addressing the bogus question. "Wait a minute man, do what with that person?" is the right type of "I'm lost" answer, not "talk to me about chicken alfredo".
My first Journal Entry ever, in 8 years! http://slashdot.org/journal/365947/aphelion-scifi-fantasy-horror-poetry-webzine
Would it be worthwhile to start running a test to see whether machines can be programmed to determine reliably whether the machine itself is in communication with a human or another machine? Would this "reverse-Turing" test more reliably set the bar for distinguishing artificial intelligence (a computer that can usually tell if it is talking to a human or another computer) from mere computation, than the current approach? What does it say about human beings who, themselves, cannot reliably pass the reverse-Turing test?
Me: No, I mean, have you been to Pripyat?
Chatbot: Oh, nope. I guess is a nice place.
Not at all, my friend. Not at all.
Gamingmuseum.com: Give your 3D accelerator a rest.
"has UR" (variant spelling "haz UR") is lol-speak for "has your" or "have your".
For crying out loud, I asked "How old are you?" It said, as expected, "Thirteen".
But when I said "When is your birthday?" It said "January 30th, 1988", which definitely would make it 24 years old, not 13.
I thought one of the first rules you teach kids about using the internet is to not disclose any personal information ;-) Remember that some people (not just kids) sign up to things online saying they were born on 1st Janurary 1900 or a similar default date.
Me: What is the meaning of life?
Eugene: Life is meaningless... Though not too much people commit suicide. Seems strange a bit.
He has the teenager's negative view on life down.
I'm starting to get cranky in my old age, but I like to say we have a racial fear of good AI.
I'm no spring chicken myslef, but it isn't good AI that I fear, it's idiots actually believing a computer can think. By the time my daughter gets to my age, there may be a law giving machines rights. Which would be insanely stupid, since bot-style AI isn't intelligence; the only intelligence is the programmer's. There are a lot of facts in the Encyclopedia Britannica, but nobody would argue that a book is intelligent, just the book's authors. People somehow anthropomorphise computers, making wikipedia sentient in their minds.
I don't fear AI, I fear what the fools and idiots who don't understand computers (many, many here at slashdot who've read way too much SF) and think a machine will ever be sentient.
If we do create a sentient machine, it wil be chemical, not electronic. Logic gates are nothing like neurons and synapses.
Free Martian Whores!
Me: Are you a bot?
Eugene: Yes, I'm a machine. Have you seen "Terminator"? It was about me. But that faint guy who played me was just a weak parody of my strong and magnificent metallic body!
It stopped working right after I started talking about my friend Little Bobby Tables. :(
When someone says, "Any fool can see
Computers per se can not think. That is nobvious. A program can. The problem here is that AI is stuck with definitions what AI is that where defined 60 years ago.
Simple question: what is a chat bot/AI doing while it is waiting for your next question? Nothing, I assume (perhaps downloading some stuff into its "knowledge base". Certainly it is not thinking about world supremancy. What are you doing whlie you are waiting for something? Daydreaming? Thinking about a hot girl? Thinking about a tough problem?
As far as I'm concerned a thinking machine is super simple to implement. Instead of trying to make it an "AI" you only need to program a thinker.
That does not necessaryly qualify it to pass a turing test, but it will be able to think and solve problems. At least it will be honestly able to answer the question: "What are you thinking about?"
Cost free eBook I read (by iBook/Kobo/Amazon/ObookO/Gutenberg etc.): "The Green Odyssey" by Philip Jose Farmer.
With all due respect to Alan Turing, that test has become counter-productive. He himself called it a game. The second word in AI is intelligence not human. Humans we've got plenty of, intelligence, not so much. What good is a deceptive computer? A calculator that gives incorrect answers. Until someone makes a Caprica-shaped robot body, I'll keep Michahonest, smart and friendly.
Fifty years of Yippie! 1968-2018
After I repeated a question about whether he called his grandpa by this first name, or simply "grandpa":
"My grandpa decided that he is a beaver right after my dad told him 10 years ago that he wants to be a gynecologist. Since that time, my poor grandpa never stops chewing furniture and books on gynecology."
My point was that there was a greater chance of speaking with a human than with a bot. That might alter your perception.
"Science can amuse and fascinate us all, but it is engineering that changes the world. " - Asimov.
You made me laugh.
Your analysis is correct. The thing to note here is that these kids seem to be announcing things and speaking their minds. They're not engaging in back and forth questioning, which is what we would assume would happen in a conversation during a turing test.
"Science can amuse and fascinate us all, but it is engineering that changes the world. " - Asimov.