{"id":4435,"date":"2022-07-08T10:00:00","date_gmt":"2022-07-08T10:00:00","guid":{"rendered":"https:\/\/modernsciences.org\/staging\/4414\/?p=4435"},"modified":"2022-06-20T08:29:17","modified_gmt":"2022-06-20T08:29:17","slug":"we-taught-an-ai-to-impersonate-shakespeare-and-oscar-wilde-heres-what-it-revealed-about-sentience","status":"publish","type":"post","link":"https:\/\/modernsciences.org\/staging\/4414\/we-taught-an-ai-to-impersonate-shakespeare-and-oscar-wilde-heres-what-it-revealed-about-sentience\/","title":{"rendered":"We taught an AI to impersonate Shakespeare and Oscar Wilde \u2013 here\u2019s what it revealed about sentience"},"content":{"rendered":"\n
\n \n
\n Shakespeare AI.\n iku4<\/a><\/span>\n <\/figcaption>\n <\/figure>\n\nAlex Connock<\/a>, University of Oxford<\/a><\/em> and Professor Andrew Stephen<\/a>, University of Oxford<\/a><\/em><\/span>\n\n
\n

Nay, nay, I say! This cannot be,<\/p>\n\n

That machines should e’er surpass our art.<\/p>\n\n

We are the masters, them the slaves,<\/p>\n\n

And thus it ever shall be so!<\/p>\n\n

They learn, \u2018tis true, but they learn<\/p>\n\n

Only what we bid them learn, no more.<\/p>\n\n

They cannot understand the heart<\/p>\n\n

Or beauty of our words, you see.<\/p>\n\n

So let us not give in to these<\/p>\n\n

Machines – they\u2019ll never be as good<\/p>\n\n

As we are at creating art.<\/p>\n<\/blockquote>\n\n

If that reads a little like Shakespeare defending humans\u2019 innate superiority over artificial intelligence hundreds of years ahead of his time, it\u2019s not. <\/p>\n\n

But it is something almost as far out: an AI system trained to express itself like the bard. The AI assimilated his style and perspective by ingesting his plays \u2013 educating itself to give an opinion on AI creativity in iambic pentameter.<\/p>\n\n

\u201cShakespeare\u201d was speaking as part of a debate held in the University of Oxford Union featuring AI versions of classic writers and literary characters. <\/p>\n\n

The motion was: \u201cThis house believes most of the world\u2019s content will soon be created by AI.\u201d<\/p>\n\n

This was a follow-up to another special AI debate that we described in The Conversation<\/a> a few months ago. Whereas that one featured an AI in its \u201cown character\u201d discussing the ethics of its technology, this time, by taking on different personas, we were able to explore this subject from a very different angle. The timely question is whether human-created content will soon be overwhelmed by the \u201csynthetic.\u201d <\/p>\n\n

Other synthetic contributors included Mrs Bennet from Jane Austen\u2019s Pride and Prejudice (1813); Winston Churchill, with a rousing parliamentary speech; and Oscar Wilde, improvising a previously unknown AI-themed scene from The Importance of Being Earnest (1895):<\/p>\n\n

\n

LADY BRACKNELL: I really cannot see what you are all making such a fuss about. It is perfectly simple. The world\u2019s content will soon be created by AI and there is nothing that can be done about it.<\/p>\n\n

GWENDOLYN: But Mama, you cannot be serious!<\/p>\n<\/blockquote>\n\n

The power of NLP<\/h2>\n\n

This creation deploys a technology known as natural language processing (NLP), in which a computer can be \u201ctrained\u201d on millions of pages of classic texts and other online content to interact with a human user \u2013 either by prompt or voice recognition. Various AIs like these have been created.<\/p>\n\n

The one we used was in the same broad category as LaMDA, an NLP owned by Google that just made headlines<\/a> after one of its software engineers claimed it was sentient. Google denies this claim and has suspended the engineer for breaching commercial confidentiality.<\/p>\n\n

The engineer\u2019s claims seem questionable, because there is little evidence that AI has achieved sentience as yet, or perhaps even ever will. But certainly AIs are already able to replicate everything from financial news reports<\/a> to synthetic Nirvana songs<\/a>, Rembrandts<\/a> and Fellini productions<\/a>. <\/p>\n\n

We have seen an AI producing images in the style of a Mughal painting<\/a> of a computer trying to persuade a crowd of wise men that it has become sentient, and deep fakes of rapper Kendrick Lamar as OJ Simpson<\/a>. Synthetic human faces are being created that we trust more than<\/a> real ones. Clearly the potential for disinformation in this space is substantial.<\/p>\n\n

Replicating bias<\/h2>\n\n

To train our \u201cwriters\u201d, we worked with AI practitioners Marina Petrova and Bruce Amick at New York agency Intentful. They trained the AI to sound exactly like the individuals whose style they were mimicking, using some 100,000 words for each that were available in the public domain.<\/p>\n\n

In our debate, we wanted to see how credibly AIs could replicate the creative text of the past, and what its outputs would be when considering its own creativity. Even great human artists concede their processing of the \u201ctraining data\u201d of their forebears. As Picasso said: \u201cGood artists copy, great artists steal.\u201d<\/p>\n\n

When we asked the Jane Austen AI to adopt the style of Mrs Bennet from Pride and Prejudice, it fascinatingly (if depressingly) picked up the gender stereotyping from the original work:<\/p>\n\n

\n

MRS BENNET: As many of you know, my husband and I have five wonderful daughters. And as any good mother would, I want nothing more than for them to be happy and successful in life.<\/p>\n\n

But in order to be happy and successful, they need to find good husbands. And in order to find good husbands, they need to be attractive to potential suitors.<\/p>\n<\/blockquote>\n\n

This was a clear reminder, as many AI developers have discovered, that bias in the training data will produce bias<\/a> in the output.<\/p>\n\n

We asked the Oscar Wilde AI to write \u201ca play in the style of Oscar Wilde, where characters are discussing whether most of the world\u2019s content will soon be created by AI\u201d. We didn\u2019t specify the play or characters, but the AI defaulted to the classic cast of Algernon, Gwendolyn and Lady Bracknell from The Importance of Being Earnest. It also invented a new character – Sir Richard. (There is a Sir Robert in Wilde\u2019s work, but in An Ideal Husband.)<\/p>\n\n

As for the AI Shakespeare, it learned the vernacular of his plays:<\/p>\n\n

\n

When machines do the work that we\u2019re assigned<\/p>\n\n

And create the content that does abound<\/p>\n\n

We\u2019ll all be free to do the things we love<\/p>\n\n

And leave the dreary work to be done by glove.<\/p>\n<\/blockquote>\n\n

Interestingly, it appears to have looked for a synonym of \u201chand\u201d to rhyme with \u201clove\u201d, and opted for the metaphorical \u201cglove\u201d. <\/p>\n\n

\n \"Robot<\/a>\n
\n \u2018Shall I compare thee to a balmy motherboard?\u2019<\/span>\n Koya979<\/a><\/span>\n <\/figcaption>\n <\/figure>\n\n

When we put the AI Shakespeare in opposition to the motion, it found an equally poetic way to justify the human intervention:<\/p>\n\n

\n

I will not soon concede that AI will best<\/p>\n\n

The works of humans – \u2018tis not to be thought!<\/p>\n\n

We have the advantage, you see, in that we<\/p>\n\n

Can reason and create, while machines but serve<\/p>\n\n

Our bidding. They may learn, it is true, but they<\/p>\n\n

Are not alive as we are, and so their works<\/p>\n\n

Will always be inferior to ours. Let us<\/p>\n\n

Not cede our dominance of the arts to these<\/p>\n\n

Unthinking machines – they cannot understand<\/p>\n\n

The beauty and the meaning of our words.<\/p>\n<\/blockquote>\n\n

Meanwhile, the AI Churchill stressed the imperative of the hour:<\/p>\n\n

\n

My fellow citizens, I stand before you today to talk about a grave threat to our society. I\u2019m talking about the threat of artificial intelligence \u2026<\/p>\n\n

We must resist this threat. We must fight back. We must stand up for our right to think for ourselves. We must defend our right to control our own minds.<\/p>\n<\/blockquote>\n\n

\u201cChurchill\u201d then pre-emptively neutralised the opposition\u2019s most potent potential arguments \u2013 in this case the accusation that he might be a Luddite \u2013 before providing a powerful, staccato conclusion:<\/p>\n\n

\n

Some say that AI will create a utopia, where all of our needs are met and we can finally live in harmony with technology. But I say that this is a fool\u2019s paradise. AI will not create a utopia, it will create a dystopia. A world where machines are in control and humans are little more than slaves.<\/p>\n<\/blockquote>\n\n

What\u2019s next?<\/h2>\n\n

This project was fun, but it is important to say what we are not saying. We are not saying this is what these great individuals would have said on this subject. We are not saying that AI is \u201cbeing creative\u201d.<\/p>\n\n

AI is merely statistically exploring training datasets. Because of its stochastic nature \u2013 involving random variables \u2013 each time you provide the same prompt, it will actually give a different answer (at one point, Shakespeare even started offering sonnets). <\/p>\n\n

\n \"Abstract<\/a>\n
\n V for variable.<\/span>\n iku4<\/a><\/span>\n <\/figcaption>\n <\/figure>\n\n

Our facsimiles of these characters are not indicative of any \u201csentience\u201d. And just as an NLP can construct a version of speech by Winston Churchill or a conversation from Mrs Bennet in Jane Austen\u2019s Pride and Prejudice, so it can construct a discussion about AI sentience with a late-night engineer.<\/p>\n\n

It is true that NLP systems are becoming effective at replicating conversation with finesse, and even quasi-intellectual engagement. But from scores of discussions with people at the major global AI companies, no one has told us they think their systems are sentient \u2013 in some cases quite the opposite. <\/p>\n\n

Debating pyrotechnics notwithstanding, AI is nowhere near the finished article yet; still a toddler at best, though growing up fast. Whether or not sentience happens, we as a society will have to grapple with these technologies and their opportunities and implications.\"The<\/p>\n\n

Alex Connock<\/a>, Fellow at S\u00e4id Business School, University of Oxford<\/a><\/em> and Professor Andrew Stephen<\/a>, L\u2019Or\u00e9al Professor of Marketing & Associate Dean of Research, S\u00e4id Business School, University of Oxford<\/a><\/em><\/span><\/p>\n\n

This article is republished from The Conversation<\/a> under a Creative Commons license. Read the original article<\/a>.<\/p>\n","protected":false},"excerpt":{"rendered":"Shakespeare AI. iku4 Alex Connock, University of Oxford and Professor Andrew Stephen, University of Oxford Nay, nay, I…\n","protected":false},"author":120,"featured_media":4436,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"nf_dc_page":"","fifu_image_url":"","fifu_image_alt":"","footnotes":""},"categories":[16],"tags":[334,474],"class_list":{"0":"post-4435","1":"post","2":"type-post","3":"status-publish","4":"format-standard","5":"has-post-thumbnail","7":"category-tech","8":"tag-artificial-intelligence","9":"tag-the-conversation","10":"cs-entry","11":"cs-video-wrap"},"aioseo_notices":[],"_links":{"self":[{"href":"https:\/\/modernsciences.org\/staging\/4414\/wp-json\/wp\/v2\/posts\/4435","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/modernsciences.org\/staging\/4414\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/modernsciences.org\/staging\/4414\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/modernsciences.org\/staging\/4414\/wp-json\/wp\/v2\/users\/120"}],"replies":[{"embeddable":true,"href":"https:\/\/modernsciences.org\/staging\/4414\/wp-json\/wp\/v2\/comments?post=4435"}],"version-history":[{"count":2,"href":"https:\/\/modernsciences.org\/staging\/4414\/wp-json\/wp\/v2\/posts\/4435\/revisions"}],"predecessor-version":[{"id":4438,"href":"https:\/\/modernsciences.org\/staging\/4414\/wp-json\/wp\/v2\/posts\/4435\/revisions\/4438"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/modernsciences.org\/staging\/4414\/wp-json\/wp\/v2\/media\/4436"}],"wp:attachment":[{"href":"https:\/\/modernsciences.org\/staging\/4414\/wp-json\/wp\/v2\/media?parent=4435"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/modernsciences.org\/staging\/4414\/wp-json\/wp\/v2\/categories?post=4435"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/modernsciences.org\/staging\/4414\/wp-json\/wp\/v2\/tags?post=4435"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}