Thanks also to my colleague Alireza.

Here’s one of the visualisations we created: the geographical distribution of collisions in 2024, classified by severity: fatal, serious, and slight.

More about the dataset:

www.data.gov.uk/dataset/cb7a...

We went through the analytics cycle: business & data understanding, data preparation, developing a predictive model and model evaluation.

I wouldn’t say it’s a bad study really. But it is misunderstood. Cognition is like a muscle. If you don’t exercise it, it will atrophy. The trick is to not offload your cognitive effort to LLMs, but to use them to augment it. This is what I teach my students.

Please could you link to it?

Thanks. I’ll check it out.

Wow! I have whiplash from how quickly that was fixed! Impressive!

Awesome. Thank you!

You can then also take them and load them up in an editor like omnigraffle and they will be fully editable.

It should work in chatgpt too though I haven’t tested it.

Bonus points if you ask to make them aesthetic and accessible.