Control for Remote Work and the AI Entry-Level Hiring Effect Nearly Disappears

Scott Galloway walks through new research that challenges the standard story about AI killing entry-level jobs. When economists put AI exposure and remote work exposure in the same statistical model, remote work explains the collapse in junior hiring and the AI effect nearly disappears. The video runs 70 seconds and compresses two studies into one argument: a working paper by Peter John Lambert of the London School of Economics and Yannick Schindler of Oxford, plus a New York Fed analysis that attributes about 64 percent of the recent rise in unemployment among young college graduates to remote work.

The Confound Hiding in the AI Narrative

Galloway opens from his podcast studio, headphones on, with a caption reading "AI ISN'T THE REAL THREAT." The setup is a measurement problem. The jobs labeled most exposed to AI are largely the same jobs that went remote: software, law, finance. On-screen captions over an animated AI graphic drive this home with the line "AI are largely the same jobs." Measured alone, AI exposure appears to explain the collapse in junior hiring. Measured alone, remote work exposure explains it too. The two variables track much of the same thing, so a single-variable analysis cannot tell you which one is doing the work.

What Happens When Both Variables Enter the Model

Lambert and Schindler tested the two explanations jointly. Their paper, "The Broken Ladder: AI, Remote Work, and Early-Career Hiring," draws on 243 million new hire records and 407 million job postings across the US, UK, Canada, and Australia from 2017 to 2025. When both variables sit in the same model, the results diverge. Jobs that can be done from home, including coding and law, show the sharpest decline in junior hiring. The B-roll matches the argument: a young man working on a laptop in a bedroom, then a screen full of code while the caption reads "jobs that can be done from home."

The Receptionist Test

The clearest evidence in the video is the falsification check. Jobs that require physical presence, including reception, show no comparable decline in junior hiring even when they use AI heavily. One frame shows a receptionist wearing a headset at a front desk with the caption "do not even when they use AI heavily." Galloway's point: if AI were the primary driver, the pattern would run the other way. Heavy AI use with in-person presence should show declines. It does not.

The New York Fed Puts a Number on It

Galloway closes with the Fed research. Economists at the New York Fed, in a Liberty Street Economics post by Natalia Emanuel, Emma Harrington, and Amanda Pallais, estimate that remote work can explain 64 percent of the recent increase in unemployment among young college graduates. Their timing argument matters as much as the number. The rise in youth unemployment predates the rapid diffusion of generative AI, which implicates remote work as the larger factor. Their proposed mechanism is practical: managers hesitate to hire junior people onto distributed teams because training them from afar is harder, and remote juniors miss the desk-side feedback and shadowing that build early-career skills.

Why the Framing Matters

If companies misdiagnose the cause, they will apply the wrong fix. Blaming AI suggests waiting out a technology cycle or retraining workers on new tools. Blaming remote work suggests rethinking how distributed teams onboard and apprentice junior staff. The receptionist comparison gives hiring managers a concrete test for their own org: are the roles you stopped hiring juniors for the AI-exposed ones, or the ones that went remote?

Key Takeaways

AI exposure and remote work exposure overlap heavily in software, law, and finance, so single-variable analyses of junior hiring are confounded.
In Lambert and Schindler's joint model, remote-capable jobs show the sharpest decline in junior hiring while the AI effect attenuates toward zero.
In-person jobs like reception show no comparable decline even with heavy AI use, the opposite of what an AI-driven story predicts.
New York Fed economists estimate remote work explains 64 percent of the recent rise in unemployment among young college graduates, and the timing predates generative AI adoption.
The likely mechanism is training: distributed teams struggle to apprentice junior hires, so they hire fewer of them.