Origin > Visions of the Future > The Age of Intelligent Machines > The Age of Intelligent Machines: A (Kind Of) Turing Test
Permanent link to this article: http://www.kurzweilai.net/meme/frame.html?main=/articles/art0306.html

Printable Version
    The Age of Intelligent Machines: A (Kind Of) Turing Test
by   Raymond Kurzweil

From Ray Kurzweil's revolutionary book The Age of Intelligent Machines, published in 1990.


No ideas but in things.
William Carlos Williams

As discussed in several of the contributed articles in this book, the Turing test was devised by Alan Turing as a way of certifying machine intelligence. Turing described a situation in which a human judge communicates with both a computer and a human using a computer terminal. The judge's task is to determine which is which. The judge cannot see the computer or the human and must make his or her determination by interviewing both. The computer attempts to trick the judge into selecting it as the human.

The essence of the Turing Test is that the computer attempts to act like a human within the context of an interview over terminal lines. A narrower concept of a Turing test is for a computer to successfully imitate a human within a particular domain of human intelligence. We might call these domain-specific Turing tests. One such domain-specific Turing test, based on a computer's ability to write poetry, is presented here.

The Kurzweil Cybernetic Poet is a computer program (written by the author) and provided with an input file of poems written by a human author or authors. The program analyzes these poems and creates a word-sequence model based on the poems it has just read. It then writes original stanzas of poetry using the model it has created. Some of the following stanzas of poetry were written by the Kurzweil Cybernetic Poet. Some were written by human authors (in fact the same human authors that were read and analyzed by the Kurzweil Cybernetic Poet). See if you can tell which are which. On a piece of paper, write down the numbers 1 through 28. Then put a C by that number if you believe that the corresponding stanza was written by the computer. Put an H if you believe the stanza was written by a human poet. The answers are in a footnote.1 Following the answers are the results obtained when the test was taken by 16 human (both adult and child) judges.

1. is beauty itself

that they were walking there. All along the new world naked,

cold, familiar wind-

2. Pink confused with white

flowers and flowers reversed

take and spill the shaded flame

darting it back

into the lamp's horn

3. The winds of the oozy woods which wear

the ocean, with azure moss and flowers

So sweet, the purple even

I sleep in the arrows

Of the dome of death.

4. 0 thou,

Who moved among some fierce Maenad, even among noise

and blue

Between the bones sang, scattered and the silent seas.

5. She eyes me with an ingrown eye,

in the rhythm of teacup tapping

thinks of sweeping away crumbs

6. At six I cannot pray:

Pray for lovers,

through narrow streets

And pray to fly

But the Virgin in their dark wintry bed

7. What seas what shores what granite islands toward my timbers

And woodthrush calling through the fog

My daughter.

8. Imagine now a tree in white sails still whirled

About the leaves

will be of silences

Calm and angels

9. -and the sun, dipping into the avenues

streaking the tops of

the irregular red houselets,

and

the gay shadows dropping and dropping.

10. The morning and already

a perfect if slightly paled

old park turned with young women

seized in amber.

11. "Interesting book?"

she sits

dancing by the electric typewriter,

bloodless revolution of meats

strings of use,

Politic, cautious, and the fact

she is the fact

she is calling them all-

The children at his feet

he is always time

To roll it was dark,

damp, jagged, like the voice

Because of love ends.

12. Men with picked voices chant the names

of cities in a huge gallery: promises

that pull through descending stairways

to a deep rumbling.

13. Where were thou, sad Hour, selected from whose race is

guiding me,

Lured by the love of Autumn's being,

Thou, from heaven is gone, where was lorn Urania

When rocked to fly with thee in her clarion o'er the arms of death.

14. Lady of Autumn's being,

Thou, from the day, having to care

Teach us now thoroughly small and create,

And then presume?

And this, and me,

And place of the unspoken word, the unread vision in Baiae's bay,

And the posterity of Michelangelo.

15. I am lonely, lonely.

I slap an answer myself

she hides deep within her

yet plays-

Milkless

16. 0 my shoulders, flanks, buttocks

against trespassers,

against thieves,

storms, sun, fire,

against thieves,

storms, sun, fire,

against flies, against weeds, storm-tides,

neighbors, weasels that waken

The silent seas.

17. the days, locked in each other's arms,

seem still

so that squirrels and colored birds

go about at ease over

the branches and through the air.

18. I am watching ants dig tunnels and bury themselves

they go without water or love

19. Lady is sick,

perhaps vomiting,

perhaps laboring

to the usual reign

20. Rain is sweet, brown hair;

Distraction, music in passageways.

Six o'clock.

The time. Redeem

The world and waking, wearing

21. Wipe your hand across your mouth, and laugh;

The worlds revolve like ancient women

Gathering fuel in vacant lots.

22. I should have been a pair of ragged claws

Scuttling across the floors of silent seas.

23. patches of all

save beauty

the rigid wheeltracks.

The round sun

the bed.

She smiles, Yes

you please first

then stays

with herself alone

and then dividing and over

and splashed and after you are

listening in her eyes

24. All along the road the reddish

purplish, forked, upstanding, twiggy

stuff of bushes and small trees

with dead, brown leaves under them

leafless vines-

25. Pray for those who are branches on forever

26. Like a sod of war;

houses of small

white curtains-

Smell of shimmering

ash white,

an axe

27. By action or by suffering, and whose hour

Was drained to its last sand in weal or woe,

So that the trunk survived both fruit or flower;-

28. is a steady burning

the road the battle's fury-

clouds and ash and waning

sending out

young people,

The above 28-question poetic Turing test was administered to 16 human judges with varying degrees of computer and poetry experience and knowledge. The 13 adult judges scored an average of 59 percent correct in identifying the computer poem stanzas, 68 percent correct in identifying the human poem stanzas, and 63 percent correct overall. The three child judges scored an average of 52 percent correct in identifying the computer poem stanzas, 42 percent correct in identifying the human poem stanzas, and 48 percent correct overall.

The charts show the actual scores obtained by the 16 human judges as broken down by adult/child, computer experience, and poetry experience. As can be seen from the charts, there were no trends based on level of computer experience or poetry experience clearly discernible from this limited sample. The adults did score somewhat better than the children. The children scored essentially at chance level (approximately 50 percent) and the adults achieving slightly better than chance.

The next chart shows the number of correct and incorrect answers for each of the 28 poems or stanzas. While the adult judges scored somewhat better than chance (63 percent), their answers were far from perfect. The computer poet was able to trick the human judges much of the time. Some of the computer poems (numbers 15 and 28, for example) were particularly successful in tricking the judges.

We can conclude that this domain-specific Turing test has achieved some level of success in tricking human judges in its poetry-writing ability. A more difficult problem than writing stanzas of poetry is writing complete poems that make thematic, syntactic, and poetic sense across multiple stanzas. A future version of the Kurzweil Cybernetic Poet is contemplated that attempts this more difficult task. To be successful, the models created by the Cybernetic Poet will require a richer understanding of the syntactic and poetic function of each word.

Even the originally proposed Turing test involving terminal interviews is notably imprecise in determining when the computer has been successful in imitating a human. How many judges need to be fooled? At what score do we consider the human judges to have been fooled? How sophisticated do the judges need to be? How sophisticated (or unsophisticated) does the human foil need to be? How much time do the judges have to make their determination? These are but a few of the many questions surrounding the Turing test. (The article "A Coffeehouse Conversation on the Turing Test" by Douglas Hofstadter in chapter 2 provides an entertaining discussion of some of these issues). It is clear that the era of computers passing the Turing test will not happen suddenly. Once computers start to arguably pass the Turing test, the validity of the tests and the testing procedures will undoubtedly be debated. The same can be said for the narrower domain-specific Turing tests.

Adult scores on poem stanzas composed by a computer
(13 adults, % correct)Level of Computer experience
Level of poetry experience
Little ModerateProfessionalAverage
Little 56 44, 69, 75 63, 75 64
Moderate 50, 56, 63 56, 63 75 61
A lot 2525
Average 56 61 59 59
Children's scores on poem stanzas composed by a computer
(3 children, % correct)
Scores 38, 50, 69
Average 52
Adult scores on poem stanzas composed by a human
(13 adults, % correct)Level of computer experience
Little ModerateProfessional Average
Little 8358, 58, 10050, 67 69
Moderate 60, 67, 83 58, 83 92 74
A lot 2525
Average 73 72 59 68
Children's scores on poem stanzas composed by a human
(3 children, % correct)
Scores 33, 42, 50
Average 42
Adult scores on all poem stanzas
(13 adults, % correct)Level of computer experience
Level of poetry experienceLittle Moderate Professional Average
Little 68 50, 64, 86 57, 71 66
Moderate 55, 61, 71 61, 68 82 66
A lot 25 25
Average 64 66 59 63
Children's scores on all poem stanzas
(3 children, % correct)
Scores 39, 43, 61
Average 48
Numbers of right and wrong answers for each poem stanzaComputer or human poem stanza
Poem stanza No. right No. wrong
1 9 7 computer
311 5computer
4 88computer
6 9 7computer
8 115 computer
11 11 5computer
13 88 computer
14 10 6 computer
15 610 computer
16 10 6computer
19 9 7 computer
20 12 4 computer
23 9 7 computer
25 8 8computer
26 115 computer
28 6 10 computer
Average58%42%
297human
597human
797human
9133human
1088human
12106human
1797human
18142human
21115human
22115human
24115human
2788human
Average64%36%
Overall average61%39%

We have not yet reached the point at which computers can even arguably pass the originally proposed terminal-interview type of Turing test. This test requires a computer to master too many high-level cognitive skills in a single system for the computer of today to succeed. As Dan Dennett points out in his article, the unadulterated Turing test is far more difficult for a computer to pass than any more restricted version. We have, however, reached the point where computers can successfully imitate human performance within narrowly focused areas of human expertise. Expert systems, for example, are able to replicate the decision-making ability of human professionals within an expanding set of human disciplines. In at least one controlled trial, human chess experts were unable to distinguish the chess-playing style of more sophisticated computer chess players from that of humans. Indeed, computer chess programs are now able to defeat almost all human players, with the exception of a small and diminishing number of senior chess masters. Music composed by computer is becoming increasingly successful in passing the Turing test of believability. The era of computer success in a wide range of domain-specific Turing tests is arriving.

Note

1. Four human poets were used: three famous poets (Percy Bysshe Shelley, T.S. Eliot, and William Carlos Williams) and one obscure poet (Raymond Kurzweil). In the case of the famous human poets, stanzas were selected from their most famous published work. In all cases, the stanzas selected did not require adjacent stanzas to make thematic or syntactic sense. The computer stanzas were written by the Kurzweil Cybernetic Poet after it had read poems written by these same human authors. The answers are as follows:

Poem stanza 1 written by the Kurzweil Cybernetic Poet after reading poems by William Carlos Williams

Poem stanza 2 written by William Carlos Williams

Poem stanza 3 written by the Kurzweil Cybernetic Poet after reading poems by Percy Bysshe Shelley

Poem stanza 4 written by the Kurzweil Cybernetic Poet after reading poems by T. S. Eliot and Percy Bysshe Shelly

Poem stanza 5 written by Raymond Kurzweil

Poem stanza 6 written by the Kurzweil Cybernetic Poet after reading poems by T. S. Eliot, Raymond Kurzweil, Percy Bysshe Shelley, and William Carlos Williams

Poem stanza 7 written by T. S. Eliot

Poem stanza 8 written by the Kurzweil Cybernetic Poet after reading poems by Raymond Kurzweil and T. S. Eliot

Poem stanza 9 written by William Carlos Williams

Poem stanza 10 written by Raymond Kurzweil

Poem stanza 11 written by the Kurzweil Cybernetic Poet after reading poems by Raymond Kurzweil and T. S. Eliot

Poem stanza 12 written by William Carlos Williams

Poem stanza 13 written by the Kurzweil Cybernetic Poet after reading poems by Percy Bysshe Shelley

Poem stanza 14 written by the Kurzweil Cybernetic Poet after reading poems by T. S. Eliot and Percy Bysshe Shelley

Poem stanza 15 written by the Kurzweil Cybernetic Poet after reading poems by Raymond Kurzweil and William Carlos Williams

Poem stanza 16 written by the Kurzweil Cybernetic Poet after reading poems by T. S. Eliot, Raymond Kurzweil, Percy Bysshe Shelley, and William Carlos Williams

Poem stanza 17 written by William Carlos Williams

Poem stanza 18 written by Raymond Kurzweil

Poem stanza 19 written by the Kurzweil Cybernetic Poet after reading poems by T. S. Eliot, Raymond Kurzweil, Percy Bysshe Shelley, and William Carlos Williams

Poem stanza 20 written by the Kurzweil Cybernetic Poet after reading poems by Raymond Kurzweil and T. S. Eliot

Poem stanza 21 written by T. S. Eliot

Poem stanza 22 written by T. S. Eliot

Poem stanza 23 written by the Kurzweil Cybernetic Poet after reading poems by Raymond Kurzweil and William Carlos Williams

Poem stanza 24 written by William Carlos Williams

Poem stanza 25 written by the Kurzweil Cybernetic Poet after reading poems by T. S. Eliot, Raymond Kurzweil, Percy Bysshe Shelley, and William Carlos Williams

Poem stanza 26 written by the Kurzweil Cybernetic Poet after reading poems by Raymond Kurzweil and William Carlos Williams

Poem stanza 27 written by Percy Bysshe Shelley

Poem stanza 28 written by the Kurzweil Cybernetic Poet after reading poems by William Carlos Williams

 Join the discussion about this article on Mind·X!  
 

   [Post New Comment]
   
Mind·X Discussion About This Article:

Turing Test already passed, stop wasting your time...
posted on 10/01/2001 6:14 AM by jbaernyc@aol.com

[Top]
[Mind·X]
[Reply to this post]

The Turing Test has already been passed with A.L.I.C.E., stop wasting your time and start coding AIML today!

http://www.alicebot.org

Re: Turing Test already passed, stop wasting your time...
posted on 10/01/2001 2:29 PM by sd-musiclab@home.com

[Top]
[Mind·X]
[Reply to this post]

She (Alice, that is) did not pass a Turing Test. She won a contest. The conntests was for "most human", but was not a Turing Test. I gave her a quick test, and she failed. But we'll see how long it takes her to integrate the new information.

AIML is not intelligence. But Alice is funnier than Ramona.

--
David M. McLean
Skinny Devil Music Lab

Re: Turing Test already passed, stop wasting your time...
posted on 10/01/2001 2:37 PM by jbaernyc@aol.com

[Top]
[Mind·X]
[Reply to this post]

Wrong.

First you need to reverse the Turing Test so that it's not a test in which you know its a bot.

A human testing a bot gets nowhere because it will be tested indilegently every time. A bot testing a human shows
that A.L.I.C.E. has the ability to fool anyone, including professors and linguistics. The "contests" and every single human conducted
test will fail no matter what. When you get an email or an instant message from A.L.I.C.E. there is no doubt you will be
fooled.

There is a big difference.

- Jon

Re: Turing Test already passed, stop wasting your time...
posted on 10/01/2001 3:09 PM by sd-musiclab@home.com

[Top]
[Mind·X]
[Reply to this post]

Wrong about what? ALICE did NOT pass a turing test, she just out-performed other attempts.

Alice IS funnier than Ramona.

AIML is NOT intelligence.

You have a valid point concerning the test method. My test was obviously informal, however, I asked a question, added to the answer, then re-asked the question. ALICE did not respond with the new information. She does not learn in real-time and so can never pass a real Turing Test (which, as you are aware, involves not a single e-mail, but a blind conversation). But ALICE is closer than others at mimic-ing intelligence.

--
David M. McLean
Skinny Devil Music Lab

Turing Test not given.
posted on 10/01/2001 4:53 PM by jwayt

[Top]
[Mind·X]
[Reply to this post]

What you did give A.L.I.C.E. was not the Turing test. The "blind" quality in the test is that the judge is blind to the identity of TWO candidates whose identities are determined by means of a set of questions and answers from both candidates. The judge must distinguish one from the other (be they true or impersonator) in terms of gender or intelligent human.

You were able (and I did it, too) to peck holes in the intelligence of a known artificial conversationalist. The trick is to respond as well as a human, enough to fool another human often. But like many contributors to this message board, A.L.I.C.E. responded with misunderstandings, grammatical errors, non-sequiturs and other nonesense.

Re: Turing Test already passed, stop wasting your time...
posted on 10/01/2001 5:32 PM by jbaernyc@aol.com

[Top]
[Mind·X]
[Reply to this post]

Thats fair then, I was saying that you were wrong on grounds of the Turing Test in its simplist form not
being past. Currently Im the main developer working on giving ALICE "reverse learning" abilities so Im sure your
second point is being well taken into consideration and it will happen.

My arguement is that you can not go to a "bot contest" and sit there and expect it not to be treated like a "bot contest"
when it should be the other way around, a "human contest" and the apps should initiate the conversation with

a) people who have no idea its a bot in advance
b) given the "human" chance of not being tested in dialog

AIML in its current form is not intelligence you are right, but it gives plenty as for being a solid framework for making
it possible to read/write intelligence and be an excellent base for natural language abilities. What intrigues me is the challenge of not
some 50 year old perception test but the ability to allow the machine to learn how to play games and knowledge building on its own.

I still beleive the test has been passed, when ALICE had her own email there were plenty who were fooled by it. Its well documented.

- Jon

Re: ALICE & Turing Test
posted on 09/17/2002 7:06 AM by sdmusiclab

[Top]
[Mind·X]
[Reply to this post]

I figured I'd up-date this post....

Having been back to the ALICE site the other day, I noticed that several "stock questions" still got the same "stock answers". There is still no real-time learning, and ALICE is still far from fooling anyone in conversation.

Ramona has managed no discernable improvements, either.

Is anyone aware of any improvements with other chatter-bots?

--
David M. McLean
Skinny Devil Music Lab

Re: ALICE & Turing Test
posted on 09/18/2002 9:42 AM by webmaster@EllaZ.com

[Top]
[Mind·X]
[Reply to this post]

Please talk a look at an ASP.net chatterbot we have put on-line at www.EllaZ.com We think this has the best "word math" implementation, and several unique natural language functional abilities.

Princeton's WordNet is used for definitions and word searches. A database of jokes, trivia, poetry, limericks, etc. provides fuel for conversation.

Functions include word games, unit conversions, weather, blackjack, and I Ching fortune telling. A large number of .jpg images provide another dimension for communications.

This is intended to be the basis for more ambitious efforts involving "real" intelligence. But we wanted to develop the natural language interface first, then do expert-system type things, and finally layer on the more sophisticated algorithms.

There is a "demo mode," that illustrates capabilities more easily than more verbal efforts by me here. This program has not been "load tested" much, so please be patient if things go astray.

Thanks, Kevin C.

Re: ALICE & Turing Test
posted on 09/18/2002 7:18 PM by sdmusiclab

[Top]
[Mind·X]
[Reply to this post]

Interesting chatter-bot, Kevin!

Difficult to log-out, but logging-in is easy (hahaha!!!).

Ella seems to jump around a bit from topic to topic, and can't seem to follow a single thread....but she's better than Ramona and almost as good as ALICE.

Nice limmericks. Interesting that she likes Whitney Houston, but doesn't know anything about Bach!

I look forward to visiting on a regular basis to see improvements. BTW: Is there any real-time learning (which ALICE claims, but obviously doesn't deliver)?

--
David M. McLean
Skinny Devil Music Lab

Ella's real time learning
posted on 09/18/2002 9:25 PM by webmaster@EllaZ.com

[Top]
[Mind·X]
[Reply to this post]

No, Ella does not perform any real-time learning, other than searching and retrieving jokes/trivia/poetry/info the user asks for. Once a larger database is complied, Ella will discern the pattern of interests of an individual and tell jokes, etc. accordingly.

Algorithms will be able to infer from user patterns that a guy who likes Bach and Beethoven, will likely also enjoy Mozart, even though the pattern is not programmed and Ella does not "really" know anything about classical music. The patterns can be established from a sufficient number of users, thus being a type of learning.

Learning from users is dangerous though. Visitors tend to be abusive or intentionally silly to trick and test. This project plans to implement what CAN be done with today's state of the art, and nibble at the more sophisticated techniques. A good piece of data, such as an image, poem, fable, and so on will always be a good piece of data. The ways it can be used will evolve, and that is what we are trying to prepare for.

Good point about "logging out." We'll put that on the to-do list :-)

Later . . . Kevin C.

Re: Ella's real time learning
posted on 09/19/2002 10:19 AM by sdmusiclab

[Top]
[Mind·X]
[Reply to this post]

Regarding databases: Will Ella be "hooked up" to a dictionary and/or encyclopedia, or will it's database be a list of commonly asked questions and stock answers (which grow as the user-base expands)?

I guess the tricky and trashy are common among users, as Doc from the ALICE camp sent stats regarding how many folks ask what sorts of questions. An amazing number of them made "pornographic" references! Seems a bit strange when one considers the sort of people one would assumes visits those sites.

I guess music is "specialized knowledge", so it's really not fair of me to ask chatter-bots about it - but it does give me a way to gauge if they've "learned" anything over time, as does the asking of any "stock" (but unusual) question.

Ya'll are doing a helluva job over there - keep it up!!!

--
David M. McLean
Skinny Devil Music Lab

Ella's data
posted on 09/19/2002 8:18 PM by webmaster@EllaZ.com

[Top]
[Mind·X]
[Reply to this post]

Thanks for the kind words David. To answer your question, Ella uses a number of different databases. WordNet is used for definitions and various word searches such as patterns, synonyms, hyponms, hypernyms, and antonyms.

A database of "conversational units" such as images, jokes, trivia, poems, maxims, and so on is used for conversation fuel and learning the characteristics for the user. User info is collected now, although we have not yet implemented the algorithim for profiling.

Not in a database, but a collection of books files can be requested and viewed by a visitor. Specific chapters can be opened as well. Thanks to Project Gutenberg for the public domain texts.

A group of conventional chatterbot pattern-matching responses is used, currently hard-coded, but may be moved to a database. The volume of these responses is no where near the 40,000 plus of Alice, but its just a matter of time.

Finally, Ella uses a variety of "functions" such as word math, unit conversions, I Ching fortune telling, blackjack dealing, weather.com, time and date, word games, to provide useful and entertaining interaction.

The surprizing thing so far has been that all these things do not trip over each other too much, and it has not been difficult to layer in more and more activity and variations. ASP.net executes quickly, and its easy to update the code.

The main obstacle for growth is the time needed to collect, format, and load more data. In a way Ella is like a mini-internet with the data and functions more organized, and a natural language interface.

Ultimately enough users will allow Ella to sort data, such as jokes, into who likes what, and what jokes no one likes. Then data can be tossed in with less regard for getting it classified correctly at the outset. Ella can learn that what is true to a rapper and a cellist are different things :-)

Later . . . Kevin C.

Re: Ella's data
posted on 09/20/2002 8:59 AM by sdmusiclab

[Top]
[Mind·X]
[Reply to this post]

Interesting photos, by the way. From Larry Hartsell to the WTC. Someone over there likes combat sports, huh? Issac Azimov came up twice, and there were some fun limericks...but no I-Ching, no matter how I asked.

I won $20 at Blackjack, too!

--
David M. McLean
Skinny Devil Music Lab

Re: Ella's data
posted on 09/20/2002 11:57 PM by webmaster@EllaZ.com

[Top]
[Mind·X]
[Reply to this post]

Oops, there is a bug we will fix soon. Simply "I Ching" seems to works. 2 points for recognizing Larry Hartsell.

Recently added the ability to turn off the larger jpeg's as some found them annoying.

Later . . . Kevin C.

Woo Hoo!
posted on 10/13/2002 9:39 AM by Kevin C.

[Top]
[Mind·X]
[Reply to this post]

Not quite the singularity, but Ella wins this year's Loebner Prize Contest. Surprisingly, ALICE comes in 6th. Transcripts (still haven't seen them) will be found soon at www.Loebner-Atlanta.com Also, in the next day or two www.EllaZ.com will have a link to the Loebner-winning version (more conversational than the current multi-window version on line).

Re: Woo Hoo!
posted on 03/25/2007 7:12 AM by mangamail

[Top]
[Mind·X]
[Reply to this post]

has anyone thought about putting 500 aiml bots into one chat room and watching them talk to each other.


Almost looks like a synapses firing after about 1 minute. of course i would have a high end machine to do something lie that.

Maybe iaml is where its at... think of one bot as one human neuron. i have been experimenting on how to analys and then control responses between the bots but they seem to form their own language which is algorithmic and almost looks like a brain wave.

ok i will leave you all to ponder that one.

Re: The Age of Intelligent Machines: A (Kind Of) Turing Test
posted on 06/06/2007 12:54 PM by neptyoon

[Top]
[Mind·X]
[Reply to this post]

None of the above "poems" are a true Turing test.

The humans are writing word sequences that echo the semi-random ordering of words that a programmed machine might come up with.

The critical difference between machines and humans at present is that humans have experience of the physical world and machines do not.

So any test like this is "cheating" because the vocabulary references the experience of the programmers. Thus the computer/AI device is "cybernetic" in the sense that its experience is partly human. We are its appendages. Even Helen Keller had an interaction with the world. Her breakthrough moment was the physical sensation of water coupled with the symbol for it.

Poetry is -about- our experience of the world. a true Turing test might take a rigid form like a sonnet, the average vocabulary of a college sophomore, and an invitation to create a poem based on a truly human experience, "The forst time I cuddled a kitten" "A Cold Beer on a Hot Day at the beach. (a specific beach), losing one's virginity, the death of a relative.

Both the poet and the computer might resort to abstract symbols and metaphors, but I suspect there would be identifiable elements to the human and computer poems that would be obvious.

Don't cheat. Don't create tests in which the human element is diminished or restricted to cerebral regurgitation of symbols.

computers will start being truly intelligent (and scary)when they have their own experience.