Taking Too Long? Why AI Content Generation Is No Longer A Shortcut To Success

If you’d told me this time last year that I’d be writing this blog in August 2026, I would probably have laughed in your face. But it’s becoming increasingly apparent that the rumours around the death of the copywriter have been greatly exaggerated.

Before we continue: this isn’t another bitter copywriter having a pop at AI. There’s plenty of that stuff around online as it is and it’s been done to death.

I completely accept that there are perfectly valid uses for AI, especially in forming ideas and initial research (even if I very rarely use it myself). And I can’t blame anyone for wanting to explore the potential of new technology if it can make a real difference to efficiency and productivity.

But I’ve noticed a real trend over the course of the summer, both through my own work and the experience of others: the business world has finally realised that using AI to generate written content isn’t perhaps all it’s cracked up to be. 

And I don’t think it’s always for the reason everyone assumes it is.

Is it really about performance?

Many organisations have started to react to the resistance that is building among their target audiences to content that is clearly AI-generated. 

This ranges from complaints about those generic posters that seem to be ten-a-penny, to LinkedIn’s introduction of a ‘Seems like AI slop’ button — which, in just a few weeks, has been used a million times by users to flag up what they think is AI content. And with Claude starting to add watermarks to its AI output, the lines between the content of man and the content of machine are starting to become less blurred.

The natural consequence of this is that AI content won’t achieve the results that it’s intended to, whether those aims are to build brand attention or interest, or get people to move to the next stage of a customer journey.

I think many businesses may be prepared to tolerate those limitations in performance when counterbalanced against the efficiency gains behind the scenes. If it’s faster, easier and cheaper to generate content with AI than it would be to get a content writer or copywriter to do it better, then it may still make business sense.

But is that definitely the case any more?

More trouble than it’s worth?

I had an illuminating conversation with a long-standing client of mine a couple of weeks ago, where it was suggested that I might like to help refine, edit or simply write many of the bread-and-butter blogs that I’d lost to AI over the last year or two.

It was a little bit of a surprise (albeit a welcome one!) to me, and so I asked them why there was a sudden change of direction.

The answer was simple: it’s taking them too long to get AI-generated content up to a sufficient standard.

The amount of back and forth and refining of prompts to beat the copy into shape is so time-consuming, that it works out easier for them to come up with a simple draft, then hand it to me to take it the rest of the way. Or potentially in some cases, to bypass AI completely and hand the brief for me to work on from scratch like the good old days.

There are two specific caveats to mention in this case. The first is that the content involved is often quite technical and requires some fairly in-depth expertise most of the time. The second is that five years of working on a weekly newspaper when I was younger has taught me how to write quickly and accurately at the same time.

But the general point still stands: marketing professionals and account managers are losing valuable time working on AI content, time that they need to be devoting to 1001 other things. This means other work gets missed, or staff get burned out, and either of those scenarios cost money to resolve. Which is the whole point of outsourcing content work to a freelancer in the first place.

Building humans back into the loop

If you work in marketing and any of the above resonates with you, ask yourself these questions:

• Is it taking you hour after hour to refine your AI-generated output?

• Are you struggling to get other work done because of that?

• Is your content still at risk of getting flagged as AI content anyway?

• Are you still having to use human expertise (either yours or someone else’s) to refine or quality-control the content before it’s published?

If you’ve answered ‘yes’ to at least three of those, then you’ve basically found yourself in a situation where you’re using AI for the sake of it. This means it’s time to start reintegrating, or to further reintegrate, copywriters and content writers into your content creation workflows.

Just remember that it’s not a one-or-the-other decision: AI may still have an important role to play at certain stages. But depending on your content, there are two ways you can approach this:

• Human-in-the-loop: where AI is used for creation of initial drafts, which are then passed to a writer for refining, editing and creative input. This will generally work better for top-of-funnel and lighter content like blogs, where the level of AI prompting work required will be lower.

• Writer-led content: where the writer takes charge of content creation from the beginning, but AI tools are used later on for quality control and sense-checking. This will suit in-depth content like reports, whitepapers and eBooks, where human expertise can put points across in detail and AI can help tidy up those points as and where appropriate.

In the content world, AI is like having an affair. It’s been fun for a couple of years, but maybe you were best with your old writing partner all along. However, your AI dalliance has taught you some exciting stuff you can use to rekindle your old relationship and get the best of both worlds. And your old partner will love you for it.

If you’re interested in working with an experienced B2B copywriter and human-in-the-loop editor for any of the above (apart from the affair), get in touch with me today or take a look at my work. 

Email me at dan@danmccalla.co.uk, or explore my portfolio here.