It is time to decide what you want
If we outsource everything, what is there left to do?
What parts of your life do you outsource? Personally, I don’t grow my own food, make my own clothes or cut my own hair. It would take ages, and the result would be hunger, hypothermia and a big mess. Cooking, however, I’ve largely kept in-house. Close management of what to put in one’s face feels strategically important.
The economy is built around getting other people to do stuff you can’t or don’t want to do. There is a service for everything. But if the cost of these services becomes minimal, what does that mean for us?
The progress in AI over the past few months has reignited the conviction that vast swathes of office work, and indeed the economy being managed by computers, is something of an inevitability. The question is: what do we want the computers to do? If we leave the economy to answer the question, everything will be couched in terms of efficiency.
What people get from their work varies so much from person to person. Some developers enjoy the act of writing code, some largely enjoy the part which involves making something. Some people enjoy both. Some people enjoy interacting with people, some people enjoy getting stuck into a spreadsheet. I wrote this piece by hand (keyboard), because I enjoy the act of writing. Writing helps me to decide what I think and allows me to express my experiences and thoughts.
In a world where AI can do a lot of this and therefore more ‘efficiently’ - does that mean it should? Capitalism says yes. Therefore businesses, driven by a competitive market, say yes. If they could deliver the same value at a fraction of the cost, why would they not? There is already a tacit assumption amongst business leaders that they need less people ‘because AI’ and it hasn’t even reached a stage of maturity yet where this is definitely true.
The good news is we have some time to decide if that’s collectively what we want. The system we have is going to have to change but the best change happens from the decisions we make as individuals. It is up to all of us to decide what we want and make it happen.
There are lots of nuances to figure out. We have to be aware of how technology subtly shapes us from spellchecks to slides. Our tools create a box, which out of habit, we are likely to stay in. If we don’t do the hard work of thinking about how we can bring our ideas to life, technology will decide. Putting a half-baked idea into ChatGPT to see what it comes up with is feeling increasingly normal. It can be a great thought partner but it is incredibly easy to forego important thinking that won’t result in any new ideas.
In the rush to adopt technology that appears to give us an edge, we risk the usual challenges with outsourcing. We hand over the keys to a third party that doesn’t have the same context in which you operate and is therefore neither able or incentivised to achieve the outcomes you are looking for. We outsourced cooking to supermarkets, and we got sloppy lasagnes.
Historically, a real lack of thought has been given to the psychological impact of new technologies. This is one of the reasons why instant communication has undoubtedly made our lives worse. Constantly responding to the never-ending stream of requests makes us feel useful and like we are ‘doing our job’, but the constant context switching and stress of a new problem to solve has made deep work much harder.
AI could have a positive or negative impact on this. The positive is that information should be far more accessible. How many messages do you respond to that could be answered by AI that has access to the right data? The negative impact is that we go from doing at least some psychologically rewarding deep work to responding to an army of agents that are doing various tasks for us.
We have to be better than AI, as many professions are rapidly discovering. There are plenty of tasks where we should use technology - I don’t think anyone is going to argue that bookkeeping should be done by counting fingers. But we have to protect the things that humans are good at, like creativity.
Generative AI is entirely based on what has happened before, it is predictive. It is also predictive without a complete set of information. Humans have a unique ability to conjure novel ideas into existence. The challenge is that the creative process is inherently ‘inefficient’. You have to think deeply about things, play around with ideas, go in circles and consider options. AI can come up with an answer immediately, but it is rarely going to be the best one.
This doesn’t mean we have to keep pushing boxes around slide decks, or writing code one line at a time. Everyone is now a bit more senior with a tireless team to do a lot of the work. But it’s a different job, getting to the best solution by managing rather than doing. It’s a job that not everyone is equipped to do and not everyone wants to do. This is the challenge organisations now face.
There are no easy takeaways here, not least because the capabilities of this Promethean technology are still evolving. All we know for certain is that the world is going to look very different in ten years and all of us have a say in what we want it to look like. If we simply let the market decide then we risk a world that no one wants. We risk a world of sloppy lasagnes.