Back to Discover

How did Anthropic measure AI's "theoretical capabilities" in the job market?

April 1, 2026
•Ars technica
How did Anthropic measure AI's "theoretical capabilities" in the job market?

Here's something that caught my attention — Anthropic released a report suggesting AI could theoretically handle up to 80% of tasks across a wide range of jobs. Now, that graph is eye-catching, especially the blue area labeled 'theoretical capability.' It looks like AI might someday do most tasks in fields from arts and media to law and management. But here’s where it gets interesting — according to Kyle Orland writing in TechCrunch, that blue zone isn’t some crystal ball prediction. Instead, it's based on pretty outdated guesses about where AI could improve productivity, not necessarily replace humans. So, while the numbers seem wild, they’re more like optimistic estimates about AI's potential, not a forecast of imminent job takeover. As Orland points out, the real takeaway isn’t fear — it's that we're still figuring out what AI can actually do versus what we hope it might do someday. Keep an eye on how these capabilities evolve; the future’s still very much in flux.

If you follow the ongoing debate over AI's growing economic impact, you may have seen the graphic below floating around this month. It comes from an Anthropic report on the labor market impacts of AI and is meant to compare the current "observed exposure" of occupations to LLMs (in red) to the "theoretical capability" of those same LLMs (in blue) across 22 job categories.

While the current "observed exposure" area is interesting in its own right, it's the blue "theoretical capability" that jumps out. At a glance, the graph implies that LLM-based systems could perform at least 80 percent of the individual "job tasks" across a shockingly wide range of human occupations, at least theoretically. It looks like Anthropic is predicting that LLMs will eventually be able to do the vast majority of jobs in broad categories ranging from "Arts & Media" and "Office & Admin" to "Legal, Business & Finance," and even "Management."

That "theoretical AI coverage" area seems like it's destined to eat a huge swath of the US job market! Credit: Anthropic

Digging into the basis for those "theoretical capability" numbers, though, provides a much less chilling image of AI's future occupational impacts. When you drill down into the specifics, that blue field represents some outdated and heavily speculative educated guesses about where AI is likely to improve human productivity and not necessarily where it will take over for humans altogether.

Read full article

Comments

Audio Transcript

If you follow the ongoing debate over AI's growing economic impact, you may have seen the graphic below floating around this month. It comes from an Anthropic report on the labor market impacts of AI and is meant to compare the current "observed exposure" of occupations to LLMs (in red) to the "theoretical capability" of those same LLMs (in blue) across 22 job categories.

While the current "observed exposure" area is interesting in its own right, it's the blue "theoretical capability" that jumps out. At a glance, the graph implies that LLM-based systems could perform at least 80 percent of the individual "job tasks" across a shockingly wide range of human occupations, at least theoretically. It looks like Anthropic is predicting that LLMs will eventually be able to do the vast majority of jobs in broad categories ranging from "Arts & Media" and "Office & Admin" to "Legal, Business & Finance," and even "Management."

That "theoretical AI coverage" area seems like it's destined to eat a huge swath of the US job market! Credit: Anthropic

Digging into the basis for those "theoretical capability" numbers, though, provides a much less chilling image of AI's future occupational impacts. When you drill down into the specifics, that blue field represents some outdated and heavily speculative educated guesses about where AI is likely to improve human productivity and not necessarily where it will take over for humans altogether.

Read full article

Comments

View original article
0:00/0:00