How will generative AI impact the creative industries?

Some early predictions

Just five months on from the public release of ChatGPT, it’s clear that generative AI is going to have a profound impact on the creative industries.

Most of the commentary I’ve read/heard to date on the likely impact falls into one of two camps: doomsday or overly-reassuring.

The doomsday camp seem confident AI is going to render huge swathes of creative roles redundant (actor, artist, presenter, musician, writer).

The overly-reassuring camp like to point to examples of things that current generative AI models can’t do as well as humans (e.g. writing comedy scripts) and imply that will remain the case.

As so often with polarised debates, the truth probably lies somewhere between the two (or possibly somewhere else entirely).

It would be somewhat foolhardy to make predictions about something which is evolving at such a dizzying pace.

However, unlike ChatGPT (“Sorry, as an AI language model, I cannot predict future events or trends”), I’m very happy to speculate and be proved wrong.

So here are three predictions:

1.) More jobs created or evolved than destroyed

Like the Spinning Jenny, the personal computer and the Internet before it, generative AI will undoubtedly reduce the manual labour required to perform many tasks and render some current roles redundant.

However, history suggests it will also lead to the creation of new roles and the evolution of many existing roles, especially on the intersection between AI and meat space.

And, like the computer, the Internet and the smartphone, it will lower the barrier to entry to many creative endeavours (although questions remain about the cost of access to generative AI).

The explosion in publicly accessible creative output enabled by the Internet hasn’t thus far led to a diminution in the creation or appeal of professionally-produced content.

Which leads on to my second prediction…

2.) An increased premium on the humanmade

Video didn’t kill the radio star, photography didn’t kill painting, recorded music didn’t kill live performance and TV, VHS, DVD, the PVR and streaming haven’t, between them, managed to kill the movie theatre.

Each new medium has brought something new to the party (vision, fidelity, portability, choice, convenience, control) but hasn’t obliterated its antecedents.

A huge part of the enduring appeal of those ‘legacy’ media is their humanity. Ken Bruce’s PopMaster, painstakingly-applied brushstrokes, a communal viewing/moshing experience.

Yes, it will become increasingly possible to create most forms of media with minimum human involvement.

But the desire for human involvement/connection seems pretty hardwired into our psyche and invariably accompanies predominantly machine-generated experiences (see Twitch).

It’s also not just about the finished product. Digital distribution has enabled much more of the creative process to be shared; from DVD extras to pottery TikToks.

However, ascertaining the degree of human involvement in creative output is going to get much more difficult  (a ‘making of’ Instagram video will no longer be a guarantee).

Which leads to my third prediction…

3.) New ways of demonstrating provenance/authenticity

We’ve limped thus far through the Disinformation Age, relying primarily on trusted news providers (now an endangered species) to identify and call out doctored images and videos.

However, the quality and speed of creation of synthetic media has just hit warp speed (check out this demo).

There is now an even greater imperative for governments, public institutions and private companies to work together to find new ways of reliably establishing and communicating the authenticity and provenance of creative works (finally a meaningful non-crypto use for blockchain/NFTs?)

If these three predictions seem overly optimistic, I’d encourage you to watch the whole of The A.I. Dilemma (which the previously linked demo is from), which unpacks the potential societal impact of companies racing to evolve AI in public without the understanding and regulatory guardrails needed to avoid an even bigger trainwreck than social media companies’ race to maximise engagement.

The creative industries may not be at grave risk from AI, but the fabric of society potentially is.