At Carat, our mantra is to make every impression matter, but today’s editorial teams face a challenge that has fundamentally changed shape. Not too long ago, the question was whether we should integrate AI into editorial workflows. That debate is over. The question now is whether the integration is deep enough, and whether our team are using the latest technology to genuinely raise the quality of what we produce, rather than simply creating more of the same, faster.
Clients expect more content, faster turnarounds, with flawless consistency across every channel and touchpoint, and they’re right to, but the context has shifted. AI is more than just a productivity experiment – it’s the operational baseline of every competitive editorial team.
The bottleneck has changed, but it’s still there
The old pain points haven't gone away, but a new pressure has emerged alongside them, where the sheer volume of AI-generated slop flooding every channel has raised the bar for what genuinely stands out.
In the past, I've watched brilliant, creative writers spend hours on mechanical tasks when they could be (and would much rather be!) crafting emotionally charged narratives and brand voices that make a real difference. Now, there’s an additional challenge: as AI democratises basic content production, the only copy worth commissioning is that which AI alone can’t produce – copy that's led by strategic insight, genuine originality and an unmistakable brand voice.
AI is an orchestration layer, not a drafting assistant
The framing of AI as a writing assistant that’s useful for first drafts and meta descriptions undersells where we are right now. The more accurate picture is AI as an orchestration layer across the entire content lifecycle, with writers as the strategic directors of that process.
This means AI handling research, structural outlines, first drafts, technical optimisation, and content adaptation across channels, it means agentic workflows that can take a brief through to a publishable asset with minimal manual friction, and it means writers spending their time on the work that AI demonstrably cannot do. What do I mean by that? Creating genuinely novel frameworks, counterintuitive insights and brand voices that feel different from every other voice in the category. Teams that are still treating AI as a drafting shortcut are leaving most of the available efficiency (and all the creative upside) on the table.
What good integration looks like
Here’s where our editorial teams are deploying AI across workflows right now:
- Research and synthesis. AI compiles, summarises, and cross-references information at a speed no human researcher can match. Writers direct the inquiry, apply critical judgment, and build the narrative, but AI does the heavy lifting.
- Outlines, drafts and iteration. The blank page block is largely gone for teams using AI well. More importantly, iteration cycles have accelerated, with writers testing unpredictable structural and tonal approaches in the time it once took to produce a single first draft.
- Technical and GEO optimisation. Short-form technical content, such as meta descriptions, ad copy, and alt text, is now almost entirely AI-assisted, but the scope of optimisation has expanded. Generative Engine Optimisation (GEO), structuring content to be cited and synthesised by AI answer engines like ChatGPT, Perplexity, and Gemini, is now a core editorial discipline, not a mechanical afterthought.
- Content adaptation and scaling. Well-crafted and strategic copy can be adapted across social, newsletters, scripts, and short-form assets with minimal additional human effort. This is now standard practice, not an advanced capability.
The human differentiator
The true value of AI integration is clear when we consider what it protects: when writers are freed from routine production tasks, they can focus completely on the work that AI can’t replicate.... that’s having strategic, original ideas, casting experienced editorial judgement and creating brand voices that strike a chord and earn genuine audience loyalty.
Writers come to this field for storytelling and creative problem-solving, and AI can restore that balance, replacing the mechanical volume of the old model with space for the creative work that drew people to editorial careers.
The teams building authority in their categories are the ones using AI to handle production volume while investing in human talent for original thinking, expert perspectives and voices that AI-generated content can’t approximate.
Rethinking what editorial talent means
Successful integration at this stage requires more than adopting the latest tools. It calls for a clear model of what our human editorial talent is there for. Basic training should develop writers as AI orchestrators who can craft precise prompts, evaluate outputs with critical rigour, maintain brand consistency across AI-assisted workflows, and recognise where their perspective wins and where it’s non-negotiable.
Track time savings and output volume, and pay attention to quality indicators – client satisfaction, reduction in revision cycles, the performance of AI-assisted content against purely human-written benchmarks, and increasingly, brand visibility in AI-generated search responses.
Where do we go from here?
The editorial teams best positioned for the next phase are not necessarily those with the most AI tools, but those who’ve built the clearest sense of what their human talent is for. It’s impossible to ignore the reality of AI handling production volume and structural optimisation, but the irreplaceable contributions are strategic originality, genuine insight and brand voices that are unmistakably and demonstrably different from everything else.
The question is no longer whether AI will change how editorial teams work, but whether they use that change to create genuinely better copy, rather than more of the same faster. Those who invest in stronger creative thinking will be the ones who pull ahead.
