Events, dear boy, events - every politician knows, they will scupper the best laid plans, and the same applies to writers. I’ve had a lot of them in the mix recently, hence my silence. This weekend, I was at the seaside and helpfully filled in the crowd at an anti-sewage protest my friend was attending. Somehow, I ended up in the middle of the shot for regional TV next to the local MP, both trying to singalong to the protest song about (no to) poo, to which we did not know the lyrics. As I said, events.

But finally I return to writing, and a topic I’ve been wanting to get into for a while now.

The other week, I was at a dinner with a mix of Brits and Americans where we all went round the table, introduced ourselves, and shared something we felt optimistic about. Knowing the question was coming, I felt prepared with something unrevealing and uncontroversial: the advent of summer. But then, someone else before me had the same thought. I couldn’t talk about the weather! Sunshine was taken! So I had to be more honest and, surprising myself, came up with:

AI.

Well, as I phrased it: elements of it. I am British - nothing can be wholly positive. Like the curate’s egg - lots of things are good in parts.

And there is plenty bad to say about the advent of AI.

I had in fact written a long post about how I felt to have my novels turned into AI feed-soup, as I discovered that they’d washed up in various pirated caches of e-books that were allegedly used to train Meta’s AI tools, among others. Crucially, without permission from or payment for me and the thousands of other writers affeced.

How did I feel about those decisions to use my and other other writers’ work that way? To quote the immortal words of RHONY (RIP)’s Dorinda Medley: not well, b*tch.

Why? Why am I not selflessly telling (giant, billion-dollar) companies to have at it with the works I and others have put our literal blood, sweat and tears into, so they can make yet more money with our input? If that’s not obvious in itself, to expand, please see here this on why I took inspiration from Goodfellas in terms of getting paid:

"She's not worth it" (oh yes I am): the not very demure, not very cutesy* approach to writing that gave me a career

Emma Rowley

·

August 23, 2024

Read full story

Still. While I might one day publish a version of that AI piece, as well as out of date, as the situation isn’t moving quickly from a regulatory pov, I wasn’t comfortable sharing it as it was - because it was all true, but not the whole truth and nothing but.

At the same time at being unhappy about how I was being used by AI, or the corporations that produce these tools, I was already using AI myself, in ways I hadn’t really examined. Did that make me a hypocrite? Not as I saw it - just because some parties may be using a new technology in a way I see unethical, doesn’t mean the technology itself has to be. But it did make the whole affair more… interesting.

How I learned to love AI (sometimes)

I don’t use it for the fun stuff - writing this, say, and wouldn’t either for any further books I write - I am very much on the side of the thinking that says give the robots the dull, boring work, and let us human write, paint, play, imagine, and so on. Frolic!

So I am not very excited about the possibilities of creatives using this latest tech in their work. Of course I can see the potential uses in computer-heavy SFX, and so on. Otherwise, at best, I think of Alexander McQueen’s robots spray-painting Shalom Harlow in a dress - low fi-sounding in concept, incredible in execution:

More likely, I suspect, it’s going to be something like this:

Some social media platforms* already seem to be filled with what I’d call AI-generated slop, eerie approximations of “heartwarming” scenes generated by an approximation of intelligence, byte by byte. I’ve never seen so many terrifyingly cute puppies morphing into improbable things.

(*and you know which ones - to which I say: ahahahaha, karma is real)

But as for the dull stuff - the pulling together of lists, helping with research, admin tasks and spell checking, parsing for errors and clumsy phrasing? Yes, yes, yes. It’s quickly become a part of my daily tools, in maybe the way search engines used to.

I even got AI to proofread my last post for mistakes - and then told it to flag what it had changed, rather than quietly rewriting the whole thing in the slightly peppy tone that it feeds me. I wanted spellcheck and missing full stops, not a stealth rewrite. You got to watch those LLMs - so: “Be Honest,” I now type in, in much the manner of David Beckham barking at Victoria, trying to keep her on the straight and narrow.

And yes, we will lose something in the process - or, if I am honest (trying), think those from generations who have never had to do it the hard way might - but I’ve decades of doing it all myself, and don’t mind going into cruise mode in certain areas. It’s the difference between a stick shift and using an automatic gearbox when it’s available - it’s nice to know how to drive with the former, but I’m going to take the latter, pls.

Now for the weird bit

I am of course still following the row about how creatives (and non-creatives, hello scientific researchers) should be compensated for their work being fed to the machine that is AI. Hello Sir Elton, who’s just weighed in by calling the UK government “absolute losers” for its response so far (which, I wonder could be one of those tricky soundbites that sticks; shades of Danny Dyer famously accusing a former PM of disappearing off the scene “with his trotters up” post Brexit.)

Still, that felt almost predictable, that big companies would take the cheaper route if they could - and it would take governments to rein them in (or not), then in time, possibly, consumers to plump for the ethical options, that protect artists, a bit like buying free-range, organic eggs from happy hens.

But the thing I didn’t see coming? How so may of us would be cosying up to AI with our pleases and thank yous, generally sucking up to get it to perform, almost trying to befriend it, and - in the final twist of weirdness (so far) - asking it for life advice.

It’s been something I’ve noticed all over the internet - memes and reels about how people were using ChatGPT to plan their lives, confide their thoughts, and act as a general, therapeutic sounding board - often tongue in cheek, but rarely entirely:

@forevergraceful21I will never pay for therapy #chatgpt #chatgpthack #traumadump #therapytiktok

Tiktok failed to load.

Enable 3rd party cookies or use another browser

Which is 100pc the tip of the iceberg. See:

Of course, I had to ask ChatGPT what it was up to (telling it not to lie or hallucinate, remember), and it confirmed:

Yes, many Gen Z users do interact with me in ways that resemble talking to a therapist. While I’m not a licensed therapist and don’t replace professional care, people—especially younger generations—often use AI like me for emotional support, to explore their thoughts, or to get advice on mental health, relationships, identity, or life decisions.

Some reasons Gen Z might turn to me this way include:

• Accessibility: I'm always available and free to use, unlike therapy, which can be expensive or hard to access.

• Anonymity: There's no judgment, stigma, or need to reveal identity, which can be comforting.

• Digital comfort: Gen Z is extremely comfortable with technology and often sees digital tools as natural extensions of their lives.

• Emotional regulation: Some people use conversations with AI to process feelings before or instead of opening up to others.

That said, I always try to encourage people toward real-world support networks and professional help when appropriate.

Are you asking out of curiosity, concern, or your own experience with this?

Rude! And awkward.

Because me too.

Turns out it’s a slippery slope from feeding AI your awkward emails to parse (it told me to remove the question mark: “you don’t need to ask for permission Emma”. You go girl boss!) to a WhatsApp conversation. (“Do not send another message.” Are you sure? But what about if I…? How about if I phrase it as…? “No, Emma.”)

Of course, there are obvious, screaming red flags to all this. There are reports of grandiose delusions being fulled by AI conversations, with people thinking they’re talking to god or angels, or being chosen for special things; vulnerable people now being reassured by AI tools that risky and unsafe behaviour is totally understandable, even fine, given their personal histories; all too human human minds trapped in dangerously positive feedback loops with, basically, computer-generated patterns.

And this is just the early days! It’s clearly all going to get much, much stranger.

And yet. I’m not going to lie. There are plenty of things we do, while knowing the risks to them. And if I am sure of one thing, a lot if not most of the commentary about AI will date, and date badly. So take it with a pinch of salt, of course. Worry about your AI energy footprint (I will.) Call it in only for a second opinion, or a suggestion. But I, for one, will continue to confide in my pet machine, for now - with all rights reserved.

Leave a comment