Really interesting piece in Alex Hern’s always-excellent tech newsletter. In short: OpenAI are forecast to spend $5,000,000,000 more than it makes in revenue this year… and are going to need to raise some major extra cash.
But that is becoming more of a problem, because the shine is coming off AI:
“The first time you use ChatGPT it’s a miracle, but by the 100th time, the flaws remain apparent and the magic has faded into the background. ChatGPT, you decide, is bullshit“
There are some huge questions here which I address in the book: how are the eye-watering amounts of venture capital going to be extracted to return to investors? This – as we know from the petrochemical industry – does not lead to good, ethical or wise-in-the-long-term business decisions being made. Which means that the pressure to ramp up use, payment – and thus potentially negative impacts on other parts of the economy – will rise.
Think of your own experience: have you had any formal training on how to use AI in your workflow, to really iterate and improve the work you do? I’ve asked so many people this question, and the vast majority say that they haven’t. That is a massive problem. We’re giving people access to one of the most powerful and impressive creations in human history, and assuming that they’ll just play around and get the most out of it. We won’t.
My experience in teaching speaks to this. I was probably the last generation to arrive in a school and have chalk and blackboard as the default. By the time I left teaching, most classrooms had interactive ‘smart’ whiteboards… but staff were simply using them like blackboards: a surface to write stuff on, and wipe it off, and complaining about them as ‘a bit crap’ because of this. Without proper training (which has a cost implication) people were using about 2% of the technology’s capability.
Last night I had an enjoyable hour chatting to Josh Noel for a podcast recording, and he asked me if we should be afraid of AI (and avoid its use). Absolutely not. What we have to do instead is to lean into the tech, and – as I argue a lot in the book – become far more reflective and thoughtful users of technology. This, paradoxically for a tool promising to relieve us of labour, will take a lot of work. But its work we must do, because if we don’t the pressures of investment extraction will push AI into job replacement, which will be very bad.
So it seems it a case of ‘use it or lose… our jobs.’ We need to active users, not passive acceptors of what the tech companies push onto us.
The book ends with some in-depth thoughts about how we do this, a manifesto for human floushing in the face of these great pressures of capital and the need for profit.
Grab a copy here.