Articulation Premium
I am completely AI-chatbot-pilled.
I am using at least Claude, Gemini and ChatGPT literally every day, and I find myself increasingly trusting and delegating more elements of my daily cognitive work to them - both personal and professional.
One thing that has become screamingly clear to me is how important it will be in the AI era to be articulate. Accurately explaining a task you are trying to complete - or a problem you are trying to solve - results in literally a 10-100x better response than a lazy, vague command. This has probably always been the case in terms of delegating anything to anyone, but here it's different in a couple of important ways.
Firstly, the feedback loop is immediate. A poorly constructed prompt delivers an underwhelming answer straight away, rather than someone toiling away in ambiguity for days or weeks before coming back with the same underwhelming answer. The speed of response underscores in real-time - literally almost live whilst I am thinking through the task myself - that I should be clearer.
Secondly, knowing AI has effectively distilled the entirety of human knowledge and is being applied to extremely complex problems highlights an important truth: I am the problem. Not my 'dumb' intern who lacks initiative or intellect. Nope, it's all me. I need to take the time to step back, slow down, and articulate what I am trying to achieve. No more blame game.
Luckily, as with the macro theme of AI in general, the marginal cost of a poor request - in terms of time, cost and intern sweat & tears - is much lower in AI world, so the penny drops pretty quickly to steer me towards better delegation and 'management' of my AI underlings.
But it makes me wonder - are the best managers actually about to become some of the best individual contributors, by applying their excellent delegation skills to software rather than humans? Is middle management on the brink of a renaissance?