In terms of technology, generative AI is often little more than a sophisticated autocorrect. Where autocorrect predicts the next word based on statistical probability, generative AI employs the same mechanism at scale, offering the most likely output for any given prompt. Competent? Certainly. Distinguished? Rarely.
Consider this scenario. A brief arrives: create a brand communication concept for the bottled water brand Simple Sips. Feed this to a generative AI model, and it will identify the obvious associations (simplicity, purity, ease), and return something like this:
"The anti-complexity campaign: Water, without the drama. Whilst most bottled waters sell you a personality or performance-enhancing lifestyle, Simple Sips sells you water. It's a palate cleanser for a complicated world."
The output is logical. It's on-brand. It's thoroughly expected. We've reached the finish line faster and cheaper, but the work is unremarkable and devoid of culture or context.
This shouldn't be surprising. The industry’s collective experience with AI tools barely spans three years. Currently, we're all clutching hastily earned bachelor's degrees in prompting. To escape the generic output trap, we must fundamentally rethink how we integrate AI into the creative process.
Traditionally, the creative process has been largely linear. In many agencies, it is a relay race from client to strategist to creative, each interpreting the problem and passing the baton to the next.
We've mistakenly applied this same linear model to AI: "Here's my problem, figure out a solution." We wait anxiously at the finish line, hoping the machine will break records, whilst the human, who understands the business problem and is both contextually and culturally aware, is relegated to spectator.
This is why so much creative work that is produced by AI is, at best, mediocre. To produce distinctive work, AI must be woven into the entire process, from brief to output and back again, in a continuous loop of human-machine collaboration.
At Forge, we've developed a methodology we call the ‘Collaborative AI Sprint’. Rather than treating AI as a creator, we position it as a process participant. In these sprints, AI generates volume, not final answers. At the brief stage, it defines options for goals. From goals, it provides options for strategic platforms. From platforms, it ideates creative execution options.
Between each stage, we insert human breakout sessions. The team stress-tests the AI's options through voting, dismantling, and reconstructing the probable into the distinct. We use AI to reduce the friction of generating ideas, but we rely on humans to carefully curate, critique and create.
Consider the mathematics: four people in a traditional brainstorm might spend an hour generating 20 generic ideas. Generative AI delivers those same 20 in seconds. This buys the team an hour not to generate, but to refine, subvert, and polish those generic starting points into work that engages, connects and resonates.
Sometimes we act as filters, refining AI's generic output. Other times, the roles reverse: we record a rambling voice note expressing half-formed thoughts, feed it to our platform, and it returns a coherent brief. Othertimes we share personal anecdotes with a research agent, and they validate our niche experience as universal truths through data.
What, then, is the agency's future role? How do we reclaim margins and creative time in this new landscape?
If AI provides the average, the human role is to recognise what "good" and “exceptional” look like and to work with technology to produce this.
To my mind, the agency of the future will be a collection of individuals who constantly collaborate, learn, share, and iterate to sharpen their taste and become smarter about using AI to grow brand equity and boost business value.
Technology and tools will continue to accelerate the time that it takes to deliver creative work to market. But our responsibility is to ensure we improve alongside these tools. Not in our ability to prompt faster, but in our capacity to discern better.
Speed without taste is simply efficient mediocrity. And sophisticated customers don’t resonate with the bland.
The agencies that survive won't be built for speed, but for contextual and cultural intelligence. They will collaborate with AI to ever improve their craft.
About the author: Courtney Chapple is the director of artificial intelligence at Forge, an award-winning, AI-powered creative agency that reshapes how marketing campaigns are created, developed, and delivered.
About Forge: An award-winning, AI-powered creative agency, Forge reshapes how marketing campaigns are created, developed, and delivered. Led by Courtney-Lee Chapple, Forge deploys a proprietary AI system with integrated tools to enable its multidisciplinary team of strategists, creatives, and media experts to create engaging, effective advertising at a fraction of the time required by conventional agencies. The agency marries the best of artificial intelligence with the best of human creativity to generate work that resonates and is culturally relevant to audiences.
Forge's distinctive advantage lies in its ability to deliver clever campaign implementation with real-time optimisation, to build brands and boost the business's bottom line. Forge’s unique approach offers clients operational efficiency without sacrificing brand equity.
