These conversations have left many activists unsurprisingly nervous (and perhaps completely uncooperative in supporting the effort. Stop the build data centers, some of which gained steam last week). Legislators panicking is not helping. None of which Describes a coherent plan for what happens next.
Even economists who have warned That AI hasn’t cut jobs yet and there may not be a cliff ahead as a result. Coming around With the idea that it can have a unique and unprecedented impact on the way we work.
Alex Amas, based at the University of Chicago, is one of those economists. He shared two things with me when we spoke Friday morning: a blunt assessment that our tools to predict what this will look like are too rough, and a “call to arms” for economists to start collecting the kind of data that can help plan how AI will solve the workforce.
On our unusual devices: Consider the fact that any task is composed of individual tasks. Part of a real estate agent’s job, for example, is to ask clients what kind of property they want to buy. The US government has made thousands of these jobs a Massive catalog First launched in 1998 and updated regularly since then. It was this data that OpenAI researchers used in December to determine that “Exposed“One task is with AI (they found a real estate agent 28% exposed, for example). Then in February, Anthropic used this data to analyze millions of cloud conversations. What works People are actually using its AI to complete and where the two lists overlapped.
But knowing the AI exposure of tasks leads to a false sense of how much risk a given job poses, Imas says. “Exposure alone is a completely meaningless tool for predicting displacement,” he told me.
Certainly, this is ideal in the most desperate case—for a job in which literally Every work Can be done by AI without any human direction. If it costs less for an AI model to do all the work you’re paid for—that’s not a given, because reasoning models and agent AI can rack up. A coffee bill— and it may well make them, Amas says, likely to lose a job. This is the case mentioned by elevator operators from decades ago; Perhaps today’s parallel is a customer service agent who is simply triaging phone calls.
But for the majority of jobs, the matter is not that simple. And the details matter, too: some jobs are likely to have dark days ahead, but knowingly how And when This is difficult to answer when looking only at exposure.
Take writing code for example. Someone who builds premium dating apps can use AI coding tools to create, let’s say, in a day what used to take three days. This means the worker is more productive. The worker’s employer, by spending the same amount, can now produce more. So will the employer want more employees or less?