Is this band really AI? 🎶

Hi Non-Techies,

I know I promised you food pics from my NYC trip, but then the food arrived, and it looked so good, and I ate it. This just kept on happening, time and time again. I was completely powerless over it.

You’ll have to settle for a pic of me being grabbed by King Kong instead. Don’t worry, I fought him off:

Today, I’m sharing three AI things that I’ve stumbled upon over the last week that made me think of you.

The AI band making waves.

Last week, The Velvet Sundown were accused of being AI-generated.

If you’re not one of their million(!) monthly Spotify listeners, let me fill you in: The Velvet Sundown are a folky band that sings about dust, echoes, silence and driftwood.

But something felt a bit off about them. Nobody knew who they were. They managed to release two excellently produced albums in under a month. Plus, their band pics all looked a bit too…polished.

Here’s one of their songs, Dust on the Wind. The comments are pretty funny:

Earlier this week, they confirmed what many already suspected: it’s an AI project. Here’s what their Spotify profile says now:

“The Velvet Sundown is a synthetic music project guided by human creative direction, and composed, voiced, and visualized with the support of artificial intelligence.

This isn’t a trick - it’s a mirror. An ongoing artistic provocation designed to challenge the boundaries of authorship, identity, and the future of music itself in the age of AI.”

What are your thoughts? Is this “synthetic music project” an exciting new chapter in this mad AI book we’re all flicking through? Or an unwelcome guest sitting at the human-only table?

Copyright your face.

Denmark has proposed an interesting solution to the problem of AI-generated deepfakes (fake media, like images, videos or audio recordings, that pretend to be real).

The Danish government wants to change their copyright law so that everyone owns the rights to their face, body and voice. That way, if someone is the victim of a deepfake, they have a legal basis from which to take action.

It hasn’t been approved yet, but it’s apparently very popular among members of parliament and is expected to be passed. You can read the whole article by clicking the button below:

Did AI write this bit?

Grammarly has released a new, free AI-detecting tool.

AI detectors are nothing new, but Grammarly’s works a bit differently. It’s not designed for the reader; it’s designed for the writer.

Let’s imagine I’m a student writing an essay (thankfully, those days are behind me). My professor has explicitly told me that I can’t use AI. Harsh, but whatever.

Now I can ask Grammarly’s new AI tool to track my document and produce a report that verifies that the words in the document have been typed out, not copied and pasted.

When I submit my essay, I can also attach the report, so that my professor has peace of mind that my terrible mark wasn’t aided by AI, but was instead 100% a result of my own brain.

Incidentally, Grammarly also has an AI detector in the more traditional sense: you paste text into it and it’ll tell you whether AI has written it.

In a very meta move, I pasted this section into it.

Here’s what it says:

Nailed it!

By the way, we tested Grammarly’s new tool in my last weekly AI Academy live session.

If you want immediate access to a friendly (and very human) community, live AI demos and a library of learning resources and prompts, click the button below and join for as little as £25/pm.

Okay, I’m back next week to talk about the time Claude became a terrible business owner.

See you soon,

Heather