Is AI replacing instructional designers?

I've been working on a course about retail returns. Not the bit you see, the customer handing something back over a counter, but everything that happens after that. Storing it, packing it, getting it back to the supplier, sorting the refund. Done badly, this is costing the business in question £80m a year.

That's not a typo. £80 million.

AI idea generation for training course engagement

So how do you get a room full of people (or a screen full of people, working through an online course) to actually care about that number before you've even started teaching them anything?

I could have asked AI. Fed it the analysis, the audience, the objectives, and asked it to suggest an opening hook. It's entirely possible it would have come back with something like an animated news broadcast, a dramatised report on a product problem and its fallout. That's not a bad idea. It might even have been a good one.

But I didn't need to ask, because I already knew.

Specific course idea generation through AI with a track record of results

A few years ago I was hired for my instructional design services for a UK business selling children's clothing. Products like that have to meet strict safety regulations, and if they don't and something goes wrong, the consequences aren't just financial. There can be product recalls, reputational damage, and in the worst cases, legal action that ends in custodial sentences. We opened that course with an animated news report. A fictional broadcast about a fictional recall, but done properly, made to feel real. It landed hard.

I've used the same device again since, for a completely different course teaching driving instructors to be more client centred. Different subject, same job to do: get the learner's attention before you ask them to learn anything.

For the returns course, we're using the same newsreel device again. Not because AI couldn't have suggested it, but because I already knew it worked. Twice. In two industries with nothing else in common. 

That's the difference between an idea that AI could come up with and a proven method that produced real results.

This is Gagne's first event of instruction, gaining attention, and it's step one of nine for a reason. Get it wrong and you're teaching to a room that hasn't decided to listen yet. AI can tell you that too, it's not a secret, it's a theory anyone can look up. 

AI training course ideas, proven track record and adaptability?

What AI can't tell you is which specific idea, out of all the ones that fit the theory, has actually worked before and how to bend it to fit a new industry, a new audience, a new brief.

That's the bit that doesn't come from a prompt. It comes from having sat in enough of these projects to know what's worked before, what hasn't, and where I can borrow an idea from one industry and reshape it for another. This is my personal training toolkit, built up over many years thanks to these unique experiences. 

I can't tell you which businesses these were, NDAs being what they are, but that's rather the point. The value isn't in the specific project. It's in having a career's worth of them to draw from, and knowing which one to reach for.

AI can't replace our experience bank

So, is AI replacing instructional designers? It might come up with some of the same ideas. It won't come with the experience bank that tells you which ones are worth trusting.

What's in your experience bank? Let me know; I'd love to know what you keep going back to.

I’m Lisa, a human freelance instructional designer with over 20 years of experience. I love technology, but I think having worked in the industry for this long gives me insights AI just doesn’t have. If that sounds good, let’s chat about designing your next training course.