Are you an LLM?
Are you an LLM?
That is the surprising question I found myself thinking about this morning as I exercised, listening to a podcast. At first it struck me as a stupid question.
But the more I thought about it, the more I realised the point behind the provocation.
We all operate daily within immersive societal norms that make us predictable, and we do not even notice the habits we follow.
Take this conversation I had at the office this morning:
“How are you doing?”
“Oh I’m fine, but I’ve been quite busy.”
“Yes, I’ve been busy too, but I’m looking forward to a rest when I go on holiday in a couple of weeks. I’ve just got a few things I need to finish first.”
That conversation could have happened 100 times over this morning, between different people, within a street or two of where I work.
That is because it is an entirely predictable conversation. It took next to no cognitive effort from either party, and yet any answer from me about what is actually going on in my day, or in my life at large, would have been more than the questioner probably wanted. They were busy being busy and predictable too.
If most of our lives are this predictable, then perhaps we can forget AGI, artificial general intelligence, which may one day apparently be conscious and a million times smarter than us all, and remember that the unconscious large language models we use today could have come up with that conversation while being entirely driven by predictable mathematics.
Do I really use my uniquely human skills at all? Or am I as predictable as a machine most of the time?
If I am not performing critical reasoning, coming up with creative solutions, or stopping to enjoy the cloud formations in the sky over my head before ducking into the office, then am I even using the fast supercomputer in my head, or any of my senses?
If we attend to the world around us like an LLM, then it is reasonable to assume that we have already been matched at the level of daily human achievement that so many of us seem to settle for.
Perhaps that is the easiest frame of reference when we are trying to see where the jobs will go.
I prepare a spreadsheet every week for a management meeting. At that meeting, my team and I go through the same performance indicators derived from the same sources.
Do we take our conscious skills for granted? Do we use our creativity? Do we use our human connection with others so that we can create something exciting and different each time we show up in our work?
Many professional environments reward repeatability. That is part of what it means to be reliable. But when people do things for the first time, when they create a product, shape a marketing campaign, write in a genuinely new voice, or paint in a new style, they are using something else. They are using the creative skills that many of us have had trained out of us since we were small. Those are the things that still stand out against the AI models.
A lot of these thoughts about how to stand out from the LLMs point towards the sort of work that entrepreneurs do every day. And the best entrepreneurs are learning that they do not need to wait for AGI, because if the frontier models available today have an IQ of 100, then as a companion they can raise the entrepreneurial IQ to a game-changing 200, provided the user maintains their agency and brings the ideas, the creativity, and the connection to the market they operate in, so that the resulting work is a co-creation. They are building something fresh and new. So much more than an LLM.