The other day I posted another one of my “hot takes” tweets about what can loosely be called “the future of work”:
A few friends asked me what I had prompted me to say this, and the short answer is that I don’t know. By this I mean that most of my tweets are scheduled weeks in advance, and I don’t recall the proximate cause of what made me think of this situation in these terms. But honestly there is not a day that passes by where a new cool tool isn’t released that simplifies some aspect of creative/professional work.
As I said in the followup to this tweet, this is the reality of what is happening right now, and not something that most of us can have much of an impact on. The genie is out of the bottle and there is no way of putting him back in. Whether we like it or not, most of us knowledge workers (myself included) will be impacted in one way or another in the upcoming months and years. My frustration is that most leaders who are aware of this are not doing nearly enough to alert us and prepare us for what is coming. Most of the high-level public discourse is focused on, real or imaginary, “AI Doom” scenarios, and attempts to forestall it. However, much more urgent and pressing issue of how to prepare the skilled workforce for the future where skills are easily “offshored” to the machines, is nearly never mentioned.
There are many reasons why we are not having these conversations. I think that one of the major problems is that we still have very loose understanding of the relationship between skills, tasks, work, and employment. In the labor economy with very low unemployment rate, (relatively) easily adjustable careers, and a fair amount of complacency with the “way things have always been done”, there is not much pressure on the stakeholders to do any major adjustments. So what if you have spent four years of your life on a degree with no marketable skills when you can spend four decades acquiring those kinds of skills with on-the-job training? So what if the job interview tests you for skills that are completely irrelevant for the kind of work that you’ll be asked to do? So what if the job only serves as a resume padding line on your resume for the next job that you really want? Up to now we have been able to dismiss all those concerns, because frankly we could afford to. We are rapidly approaching the point where that will no longer be the case.
It seems to me that one useful skill for employees to develop is to learn how to work with AI tools. Whether they end up being a complement to, or replacement for, one's job, the tools are here, they're only going to become more powerful, and those who understand how to use them will be better-positioned than those who stick their head in the sand.
The implications of this spooks me a lot, it has kept me awake many nights this year. The most peace i've gotten this year was from convicing/fooling myself back into the belief i had when i first tried out GPT3 in 2020.. that it'll "stall out, surely you can't just keep using the same methods, the data must have some diminishing return at some point..."
But it downright terrifies me how, when i look around, everyone seems to be oblivious to it (including a lot of "knowledge workers").
You are right, we are not ready for this, socially or institutionally.