There is always a good story at the heart of the best advertising. But Gen AI tools flatten the background narrative through black box explanations as to how the Ad was created. The results can be impressive but consider the behind the scenes on any generated film to a real one. If you consider the location, crew and actors, the problem is obvious. And yet it would be absurd to do in camera what is easily done with Gen AI.
This is the new paradoxical reality we are in, which is leading to a massive transformation of the creative sector. And a collapse of ambition toward high quality effective stories that make brands shine and ultimately sell products.
We need to understand when to use synthetic creative tools and when not. And when the work we are doing should have been automated and when it should not. And even more important which creative stories only work if there is a human in the picture. And even which are the stories we can unlock today? New stories to move and inspire us like the classic advertisements of yore.
I spent the last year working on a human-creative map of what AI can’t do, and in many cases, will never be able to do. The idea being that these same areas are places in which human creative work is not only safe, but necessary. We need to work on stories that are AI-safe with as much effort as creating the next automated generative pipeline.
I hope this map serves as a guide to creatives to areas that are effectively impossible for synthetic intelligence. And that they realise that stories in these areas can be lasting treasures, that we should shift attention to. All the while leaving much of what can be industrially created, to our tools.
I use a generated image as my overview map. I created it with a few days of prompting effort. I tried to create a combination of symbolic elements that would help me remember them. I gave the image a renaissance vibe. I think new creative directors will be more generalist scientists like Galileo or Leonardo, with constant access to new instruments. They will discover new possibilities but also be aware of the difference between nature and tools. Also note that the image has some elements which derive from religious symbolism. They are not intended as a spiritual comment, and I hope no one misunderstands this intent or finds it offensive.
The areas I look at are abstract and not guaranteed to be AI-free. But the map I share is a guide of where to look for human exclusive stories. Also I am not intending to judge the use of Gen AI negatively. Rather it is a tool that should have its place. At the same time I think it is important for us to understand where we get to its limits.
Here is my legend starting from the top of the picture and going down, and as I go through aspects they are not completely separate from each other nor am I able to fully describe them in brief as some are huge areas:
1. The Scales represent Ethics and Law
AI can’t sign contracts, and cannot fully solve ethical problems because it can’t take responsibility for its decisions. From a campaign point of view think of licensing talent, or politically sensitive messaging. In situations where legal aspects play a key part, there always will need to be a human involved.
One critique of this is that the human is just there to sign the contract. But great creative work can be done by weaving complex legal requirements: from licensing, IP, location permits, special talent, etc. Think of ideas that are born around and with legal questions at their core.
2. The Woman as a symbol of Embodiment and Sensory Systems
The way humans exist with their body and perception of the world is unique to us. While AI can pretend to feel a temperature, or see Elvis in a potato chip (pareidolia), it can’t experience it. From a campaign point of view, consider acoustic or optical illusions. They are inaccessible to AI in a fundamental way because it does not have the same sensory systems we have.
This area is enormous, and includes the experiential and product type work. For real humans to move through real spaces and do things with their hands with real objects is special. This will remain heavily in the domain of human creativity even as it is heavily assisted by simulations and tools.
3. The Halo and Consciousness
We are still not fully sure as to what consciousness is, but there are meaningful characteristics that define it. We know it can come and go in sleep and waking. We know we can intentionally shift our attention from a painful bump to to the music we are listening to. We know we can bend reality with too much alcohol, etc.. There are many “correlates of consciousness” that combine to define it. Synthetic intelligence is locked out of this world except through indirect reports.
Some examples of creative work that leverages consciousness would involve balancing subjective focus and visualisation. This area covers where brand values translate to customer experience . Or while trying to balance digital overlayed information with space the user is in. Or reaching out with a virtual hand shaped like a hook to grab a virtual coffee.
Consciousness is a layer in all human thinking so there are huge areas to build AI-safe ideas. But it does also require the creative to dive into research in this field.
4. The Third Eye for The Unknown and Unknowable
The unknown and unknowable are areas which can be very sensitive to beliefs humans hold. It would be irresponsible to use an AI to come up with narratives there without carefully checking them. Gen AI can mix and match different cultural references and beliefs in ways that are risky or meaningless. But a human creative director would know how to tease a connection to religion, for example, without offending.
There are people that ask AI about God. I think this is not a wise decision because we mix Reddit comments with classic religious texts of many religions. We try to find wisdom by putting everything ever written about Him in a probabilistic blender before asking. This is why this domain of creative work, I think, will always require embedded human culture and traditions.
The unknowable includes weird impossible questions. Is there a planet somewhere in the universe where the night sky stars are in the pattern of a swoosh?
5. The DNA Stranded Choker as a symbol for Genetic Influence
This is one of the most important, subtle and perhaps controversial areas we should work with. Much of our human intelligence actually is coded in our biology. Babies have a lot of innate knowledge. They know their mother’s voice, they learn about faces way too fast. Babies they can tell when an object is magically floating or not by staring for a longer time. They learn language as if there were a machine ready to accept it and so on.
Ideas in this space capture knowledge that is not even explicit because we take it for granted. Look at a sound-symbolism effect called the Bouba/Kiki effect, (which is a like Synaesthesia). Imagine a jagged design for a brand called KricketyKrick, or a rounded design for a brand called BlombedyBlomb. And then imagine the converse case.
To work in this space one has to start from knowing what the heritable characteristic is. And by having some understanding of the science one can try to weave creative ideas around the results.
6. The Child and Creativity
We can surely use AI to create new work, but it is tightly related to all prior work. When we think of new work that is “really new” then AI can’t help. If we look for work that reaches out further than past data and maintains consistency: this is human exclusive.
An easy example for this is a list of bad ideas that might become fashionable for some reason. Imagine a book where every other word is upside down, or printed with illegible defects. Or imagine with words individually spelt backwards. These sort of ideas, that recall Surrealist Art, are NEW because they are unlikely options. But a human might see to be good choices in certain specific contexts.
There are other examples of novelty that come from technical innovations or new platforms and mediums themselves. Overall, the challenge of novelty is that it isn’t new, if it also isn’t a good idea. So it is risky to look for novelty as the sole basis for ideas.
7. The Heart for Emotion/Social Connection
Emotions are complex human processes that shape human behaviour and which we do not understand. They are at the heart of motivation and reward and fundamental aspects of creativity. But, AI can’t feel emotions. So any situation whose goal is an emotive outcome, a key part of the work must involve a human.
Almost all the ideas we produce are concerned with the emotional impact of the work. But we are also used to the technical challenges involved in communication. So much so that we under estimate the nuance needed to convey an emotions. We are not used to regular testing of what things feel like; but with synthetic production, we really should.
An example idea here would be to tie brand messaging to its potential to convey a certain emotion. As an example Guinness has shared the same message for many years: “good things come for those who wait”. This message sets up an expectation for a certain style of communication that is focused on authenticity. A feeling that is much harder for a synthetic production to create reliably.
8. The Dice represent Chaos and Randomness of Nature
Nature, at its core is built up of probabilistic processes that come together in complex outcomes. These sort of phenomena are impossible to model and predict in detail. This is why, for example, it does not matter how much data we have for some phenomena. We cannot predict the radioactive decay of an atom, we can’t perfectly predict an earthquake or a storm. We have probabilities that help us get a sense something possible but not certainties.
Ideas in this space can make use of the unpredictability of natural processes to establish a domain that AI can’t take over. Most chance based games will remain fun no matter how smart the opponent because of this randomness.
9. The Black Hole as a symbol for AI Creative “Hacks”
At the base of this drawing I place a black hole: a mysterious object that absorbs any type of information. This is a symbol for “hacks” to the Gen AI tools themselves. The various Chat GPTs, Groks, Geminis and Claudes present special opportunities to play creatively around their implicit limitations. This is like the tik-tok user arguing with ChatGPT to count to one million (and it repeatedly politely declining).
There are many sub-areas in this domain, from creating inappropriate content, to tricking the AI because of its stereotypes. My favourite example are ideas around a problem that AI feels compelled to solve (like in my AI Guinness hack). The AI can’t resist following the data sets it has been trained with. If I train an AI to win a game of tic tac toe, it will struggle to lose the game, even if for a secondary benefit. In many cases, this area is about common sense. An area of human thinking which looks very easy to us, but remains the biggest unsolved problem in AI.
The Human-Creative Map Ideas Generation Game by Yates Buckley
Each one of these the areas in the human-creative map offer subcategories, and they are not all perfectly separate either. A clay sculpture can connect to an embodied/sensory idea, but it might depict an emotional scene. A collective social game might involve using LLMs to conceal messages shared between the players the AI tries to guess.
There is a huge opportunity for human exclusive ideas, but they require a bit of effort to get into. For this reason I offer a LLM game to exercise your creativity skills. I realise there is a paradox in using AI to find AI-free ideas. But if you try you will learn as much from where it fails to play well as where it succeeds. When I find mysterious wins, I look deeper for something human done by someone else, in the data, before.
Copy paste this to your favourite LLM:
=== Prompt ===
The Human-Creative Map Ideas Generation Game by Yates Buckley
Let’s play a this game. Me, human versus you assistant which take turns at being first and second players. As background we take into consideration the full description of the creative areas listed here. It is a list of categories of creative work that require human involvement.
Based on the article for the Human-Creative Map https://yatesbuckley.com/2025/09/09/human-creative-map/
- Ethics and Law (Scales) - AI can't sign contracts or take responsibility.
- Embodiment and Sensory Systems (Woman) - Unique human perception and physical experience.
- Consciousness (Halo) - Subjective awareness and intentional attention.
- The Unknown/Unknowable (Third Eye) - Sensitive spiritual domains and impossibilities.
- Genetic Influence (DNA Choker) - Hard-coded biological intelligence and intuitions.
- Creativity (Child) - Truly novel work that extends beyond training data.
- Emotion/Social Connection (Heart) - Complex emotional processes and motivation.
- Chaos and Randomness (Dice and Universe) - Unpredictable natural processes.
- AI Limitations (Black Hole) - Structural problems AI can't solve.
This framework offers guidance for creative directors navigating the AI landscape. The idea is to help know when to leverage automation and when human involvement is not just preferable but necessary.
Each turn a first player proposes one of the areas by describing the symbol. The second player must come up with an idea that fits that category. And a rating 1 to 5 of how human-exclusive this idea is. Note that 1 is very little human involvement required and 5 is impossible even for an advanced AI.
The first player can now object to the scoring. To do this the first player must propose a method AI could use to realise the idea. And propose the new lower rating. The second player can counter this objection.
At the end of the turn the second player gains added score based on the consensus rating of the idea. Or if no consensus is reached between the players the turn is skipped. The total player scores for each player is presented at the end of the turn. Also a reminder of all the categories of ideas, is presented at the end of each turn.
After 10 turns the player with the most points wins. A review of the game with each symbol and its sub-category ideas and scores are listed.
=== End ===
Reference link to sample play