You know the story: handsome but inarticulate soldier Christian falls in love with Roxane, enlists the help of honey-tongued, big-nosed Cyrano de Bergerac to woo her with love poems. (Steve Martin was great in the film.) All ends in tragedy. Something similar is happening with the rapid uptake of AI voice-to-text technology. Teachers, doctors, those on dating websites, town councillors, students submitting work, lecturers marking work, planning officers, to name only a few, are routinely relying on artificially created text to complete work assignments. To adapt Roxane: “We are falling in love with an impossibility.”
At a recent training day I was taught how to use AI speech recognition to write reports, letters and emails. A teaching assistant led the session:
English has never been my strong point and I constantly make all sorts of spelling and grammatical errors. AI has changed my life. If I have to record an incident about a pupil I now use the microphone button on Gemini, explain the incident, ask it to rewrite as a decent school incident report and wait a couple of seconds. I then cut and paste Gemini’s response and send it off.
I know and like this particular woman but she is not one of life’s natural communicators. She prefers saying “I done that” to ‘I did that’, for instance. She also has a wicked sense of humour. I read the report that she done – it was improbably immaculate.
I have written before about how AI is democratising communication for various non-professionals. I love the personal innovation of voice recognition technology. Yet, I still sense a bitter ending. Like Christian, this woman – and increasingly millions of others like her – is not being truthful. When Roxane discovers the deception she yells to Cyrano: “Who said it? Who! Whose voice did I hear? Whose words did I read? Who was I writing back to? … You wasted me… let me love an impossibility.”
What makes AI voice technology a different issue from that of mere plagiarism is the fact that the kernel of the messaging does indeed emanate from the person commissioning the AI. It’s all about the prompts. Christian did love Roxane, the teaching assistant did experience an incident with a student that needed recording. Friends and family all use the technology with pleasure. I finished off my online training by asking Claude the answers to everything and ticking the relevant boxes in another window.
With speech speeds being three times faster than typing, it makes sense to drive home from work dictating instructions. ‘Write an email to these stakeholders using the previous format explaining the next steps in the consultations’; ‘Write an objection to the new planning proposal number at this address – cite all the relevant reasons.’ My GP uses Heidi the medical scribe to write up appointments. He shouts at his computer such things as: ‘Tell Mrs Moaner that her dry skin is just aging, the waiting list for the dermatologist is a year, put some cream on it.’ This is magically turned into a polite text message with a far off promise of a dermatology appointment and a prescription for a suitable emollient. Would Mrs Moaner actually prefer a good old-fashioned doctor to shout at her?
How many modern daters are, like Roxane, duped by AI-honeyed messages from lacklustre wordsmiths? “I read, read again — grew faint for love; I was thine utterly. Each separate page was like a fluttering flower-petal, loosed from your own soul, and wafted thus to mine.” How will parents react on parents evening when they meet teachers in person and the reality falls far short of the articulate correspondence they had previously received?
Christian laments midway through the play: “I am aweary of these borrowed letters – borrowed love makings.” He decides he doesn’t need Cyrano anymore and attempts to speak of love directly to Roxane. He makes an ass of himself by only being able to say: “I love you.” His honest earnestness ought to give us pause. Is our human frailty now insufficient? Must everything, even our own speech, be pimped and polished? The wizardry that turns an inarticulate ‘brain dump’ into a well-written piece of work however is the intellectual equivalent of a boob job, false eyelashes or those grim looks-maxed men: artificial, synthetic, counterfeit.
Cyrano dies (spoiler alert) after being attacked with a log by his enemies. Like a malfunctioning computer he forgets words and speech and utters gibberish. As he dies, Cyrano sadly identifies his role – and reminds us that essentially AI is nothing, even if we are duped into believing it is everything:
It was my life, to be the prompter every one forgets! … Here lies Cyrano de Bergerac, who was everything, yet was naught.
Joanna Gray is a writer and confidence coach.
Discussion
Comments
This week across the site:
To join in with the discussion please make a donation to the Daily Sceptic.
Profanity and abuse will be removed and may lead to a permanent ban.
Excellent piece
Yes our brains are being taken over. We hardly need to think anymore.
“Yet, I still sense a bitter ending.”
I do, too.
As I pointed out on one of your previous excellent articles on AI… It may be giving people a voice, but it if it ain’t their voice….
AI is encouraging people to live increasingly in a made-up world (as if social media were not enough). But reality has a way of kicking people up the arse.
I can see it being very useful to write planning objections for instance. Don’t use it myself however.
I attended a presentation last night at a Lycee here in the French Pyrenees, with all the other parents whose children have just started the first of three years’ study for the Baccalaureate.
During the question and answer session at the end, someone asked about the availability of coding classes. The teacher who elected to answer said, words to the effect,
“I used to recommend a career in coding but now with AI doing it all I can no longer recommend it.”
I really blooming hope he was joking. Because if not, it’s people like him with these idiotic views who are the real problem.
Apart from this, we are pretty impressed with the Lycee.
I teach translation and interpreting, and even some of my colleagues have started telling students there’s not much of a future in the profession because AI will soon be better than we are. As far as I can tell – and I use AI translation tools quite a lot in my work – that is nonsense.
I have to disagree. My long career as a translator came to an end last year through a combination of clients saving money by using AI (“It’s not as good but it’s good enough”) and taxes, particularly steeply rising local taxes, taking most of any profit.
If you want a quality translation, AI/Google Translate/Deepl often really doesn’t cut the mustard. It doesn’t do subtlety, nuance, colloquialisms, local dialect…
Writing letters to contest an alleged parking infraction? Yes, good enough.
I’d connect with you to chat more in private! But DS demands that I “complete my profile” by adding either a Facebook, LinkedIn or Twitter profile, and I have none of these (nor want them). Barmy. Moderator, any thoughts?
I’m sorry to hear that. I’ve lost two clients so far to AI, but they were not really committed to quality communication as a key part of their business anyway. That was 18 months ago or so; as far as I’m aware there’s been no one since, though of course I can’t say anything useful about people who might have started employing me but chose not to without mentioning it. Being an interpreter as well helps, as the need for immediate accuracy and above all confidentiality is greater on that side of things.
Focus on the real.
I have never met an AI who can build a house.
I have never met an AI who can help my aged mother get on.the toilet when she needs to go.
I have never met an AI who can write an inspiring piece of music.
I have never met an AI who can make a piece of software that works and can fix it when it, as it often is, is found to be bug-ridden and erratic.
Stay sane, people. Stay sane.
I don’t know about the efficacy of AI generated love letters but recent research I saw reported a week or so ago, has pitted three groups of students against each other writing reports, whilst simultaneously hooking them up to brain activity monitoring.
The group that wrote their reports based on memory alone, lit up like Christmas trees, the second group who were allowed to use reference materials were similarly ‘brain active’, while the final group which briefed and used AI to write their reports were, apparently, practically brain dead through the process… a condition that was rather depressingly maintained when they were asked to discuss their findings.