Adding Insult...

While the repulsive Grok has been developed to transform images of real people into sexualised versions of them, here in the real world we seem to be in the hands of underdeveloped children untransformed by being in adult bodies.

See the White House. See Elon Musk...

With the European Commission announcing that it too will investigate Grok's appallingly insensitive possibilities, Musk has posted what appear to be two AI images of women on a beach – in one version in bikinis, and in the other in burqas, presumably implying that the latter is where we end up if Grok is subject to "censorship".

The images probably manage to be offensive to women and to muslims – and to intelligence (actual not artificial).

Antisocial Ignorance

Ban under-16s from social media? Yes, please...

I was recently surrounded by a swarm of nine-year-olds on public transport, all equipped with smartphones.

The one who squeezed into the seat next to me was grotesquely hunched over his screen and absorbed in a game which seemed to involve driving a car very fast down an endless street. He appeared to be oblivious to the damage he was probably doing to his body as well as his mind.

He and his friend, who was jammed in standing next to me, communicated with each other and moved back and forth as if I did not exist.

If their phones were taken away, it might do them – and the rest of us – a considerable favour.

Art of Illiteracy

Frank Zappa once described rock journalism as "people who can't write interviewing people who can't speak for people who can't read".

It seems that AI is going to make that quote more widely applicable, since it enables people who can't write to do so – with neither they nor the people who read the result being aware of its shoddiness.

And as everything gets dumbed down, maybe the people interviewed won't care about their words being mixed up and made part of the AI slop being generated.

Apple Inaccuracies

After an unfortunate incident involving a water leak and my laptop, I had little choice but to buy a new Apple Mac laptop.

Like all Macs, it's extremely user-friendly and I'm generally happy with it.

But I just found that among this one's writing tools are "Make friendly" and "Make professional". I tried both on a feature I was writing and the results were shocking.

Not only did both options – in different ways – change key words and make the prose unnatural, but on top of that they changed words in quotations from the interviewee so that they were "friendly" or "professional" – and completely inaccurate.

Journalists have always had to live with accusations of making up quotes. Now we've got computers making up quotes for them.

The slippery slope is getting slippier...

Adjudicating Innocence

 

An AI judge and a court in which you are guilty until proven innocent (and that proof has to be supplied within 90 minutes)... not the latest brainwave from UK justice secretary David Lammy but a new sci-fi movie set in the near future.

It's already out in the US and due for release in the UK soon.

If you're hoping for a philosophical or ethical inquiry into the possible use of AI to both try and execute people, you will be disappointed.

But if you want an all-action, shoot'em-up whodunnit with AI thrown into the usual wham-bam mix, then it could be for you.

Let's just hope Lammy doesn't find it inspiring...

Apposite Idea

 How do you worsen a situation in which AI is being used to create sexualised images of women and children?

In the case of Grok and Elon Musk, you simply make this a pay-to-use facility.

The UK government said the move was insulting to victims and "simply turns an AI feature that allows the creation of unlawful images into a premium service".

My predictive text just tried to turn Musk into Muck… So maybe AI does sometimes come up with apposite ideas…

Aptly Incisive

 

The headline writers at the Metro have consistently outdone their tabloid rivals for months – with some excellent and meaningful puns.

This three-worder from today pretty much conveys everything that I've been banging on about in the previous two posts here.

This is human wit and incsiveness at work – not AI.