|
Issue 24
|
|
The weekly brief Employers say they’d rather see your AI skills than your diploma. In the UK, that preference is already worth a 34% pay bump.This issue checks the loudest claim in hiring right now against what employers actually do — where skills-based hiring is real, where it’s mostly branding, and what still earns a wage premium when the job posting says no degree required. November 2, 2026 · 9 min read
|
Sourced from PwC, TestGorilla, Harvard Business School & Burning Glass Institute, Dice, LinkedIn, Gallup-Lumina, WGU. | No hype We checked the loudest ‘skills over degree’ claims against hiring-outcomes research, not just the press releases. | Why now UK AI-skills wage premiums tripled in a year, and over half of job postings have already dropped their degree line. |
|
If you read nothing else More than half of employers now say they’ve dropped degree requirements from job postings, and demand for AI skills is climbing faster than almost anything else in hiring. But a Harvard Business School and Burning Glass Institute study of 316 million postings found that removing the requirement lifts non-degree hiring by only about 3.5 percentage points on average — most companies kept hiring the same people anyway. The real edge goes to employers who paired the policy with new assessments and real sponsorship, not just an edited job ad. |
|
|
Nova’s signal Dropping a degree requirement is a policy change. Building a way to actually recognize skill without one is a different, harder project — and most companies did only the first and called it the second. Nova’s call. If a job posting says ‘AI skills, no degree required,’ ask what they actually changed about how they screen — not just what they removed. Last issue tracked a regulator trying to get ahead of AI in medicine before a rule exists; this one tracks employers doing something similar with hiring, without waiting for anyone to referee it. Story 01 has the wage data behind the claim, 02 the degree requirements employers are actually dropping, 03 the study showing most of that is cosmetic, 04 the demand signals pushing AI skills onto every resume, and 05 the opposite bet some employers are quietly making instead.
|
|
This week 5 developments in the fight over what actually gets you hired.
|
|
01 / Labor Market UK workers with AI skills now earn a 34% wage premium — nearly triple what it was a year ago.PwC’s 2026 Global AI Jobs Barometer, released in June and built on Lightcast job-ad and salary data, found UK workers with AI skills earned a 34.2% wage premium in 2025, up from 11% the year before. Specialist AI job postings in the UK rose 61% to roughly 180,000 over the same period, even though the share of all UK job ads explicitly requiring AI skills reached just 2.2%, and the UK’s premium still trails the 62% average PwC measured across 27 countries, per PwC’s own release. Why it matters. This is the clearest hard number behind the ‘AI skills over diploma’ claim employers keep making — pay is already moving before most hiring policy has caught up, and it’s moving fastest for a small, specific slice of the workforce. The signal. The premium varies enormously by sector — PwC’s breakdown puts consumer-markets roles as high as 64% and government roles as low as 12% — so ‘AI skills pay more’ is really a sector-by-sector bet, not a universal rule. PwC: UK AI hiring and wages surge — 2026 AI Jobs Barometer →
|
|
02 / Hiring Practice More than half of employers say they’ve dropped the degree requirement.TestGorilla’s 2026 hiring research found 53% of employers removed degree requirements from job postings in 2025, up roughly 30% from the year before, and separate NACE survey data put the share of employers using skills-based hiring for entry-level roles at 70% in 2026, up from 65% a year earlier, per Unleash.ai’s coverage. On paper, the policy shift looks sweeping and nearly universal. Why it matters. If accurate, this is exactly the premise behind every ‘companies don’t care about your degree anymore’ headline — and it’s the claim the rest of this issue tests against what actually happens next. The signal. Separate job-posting counts put the real share even lower: one analysis found just 28% of current U.S. postings still list a degree requirement at all, down from a near-default a decade ago. Unleash.ai: employers are dropping degree requirements, TestGorilla finds →
|
|
03 / Hiring Practice A rigorous study says dropping the degree requirement barely changes who gets hired.A Harvard Business School and Burning Glass Institute analysis of 316 million U.S. job postings between 2014 and 2023 found that removing a bachelor’s degree requirement raised non-degree hiring in those same roles by only about 3.5 percentage points on average. Nearly half of the employers that dropped the requirement made no meaningful change to who they actually hired, per the researchers’ own accounting, covered by HR Executive. Why it matters. This is the gap between the policy employers announce and the hiring employers actually do — a ‘no degree required’ posting is weak evidence, on its own, that a company changed how it screens candidates. The signal. About a third of employers were genuine exceptions: ‘leaders’ who paired the policy with real structural change — new assessments, internal promotion paths, executive sponsorship — saw an 18% jump in non-degree hires, and those hires stayed roughly 10 percentage points longer than their degree-holding peers. HR Executive: why dropping degree requirements isn’t actually working →
|
|
04 / Demand Signal AI fluency is now required in 75% of US tech job postings — up 178% in a year.Dice’s July 2026 Jobs Report, based on more than 7 million U.S. tech postings, found 75% of US technology openings required AI fluency in June, up 178% year-over-year, per CIO’s coverage — and Dice’s August update puts the figure even higher, at 79%, for July. LinkedIn separately reports that members who list AI literacy skills on their profile are 78% more likely to be hired and 11% more likely to be promoted, though LinkedIn hasn’t published the methodology behind that figure, so it reads as a correlation worth noting, not a measured premium. Why it matters. Whatever is happening with degrees, the demand side of this story isn’t ambiguous — AI fluency is becoming a baseline filter across far more roles than engineering alone, and showing it correlates with getting through that filter. The signal. LinkedIn also reports that postings requiring AI literacy have jumped roughly 65% in each of the last two years and now sit at more than ten times the 2023 rate — a faster climb than almost any other skill category it tracks. CIO: AI fluency becomes a baseline requirement for US tech roles →
|
|
05 / Outlook Nearly half of employers now bet AI will make the degree more valuable, not less.A Gallup-Lumina poll of 2,500 US employers in October found 48% expect AI to make college degrees more important over the next five years, versus just 26% who expect them to matter less. Separately, WGU’s 2026 Workforce Decoded report, surveying 3,128 US hiring professionals, found 60% say AI is making it harder to evaluate candidates’ real skills — and ties that difficulty to employers leaning more, not less, on degrees and certificates as a proxy. Why it matters. This directly complicates this issue’s headline claim: the same employers removing degree lines from job postings are, in meaningful numbers, telling pollsters they expect to lean on degrees more as AI makes skills themselves harder to verify. The signal. Both dynamics can be true at once: employers may keep dropping the degree requirement from the posting while quietly treating it as a tiebreaker once AI-generated resumes and interview answers make candidates harder to tell apart. WGU: AI has made real skills harder to evaluate →
|
|
Signal & chatter A few more things worth knowing.Two trackers of the same trend flatly disagree. A Lightcast analysis finds AI-skill job postings carry roughly a 28% pay premium over similar postings without them. A separate, unverified tracker of AI-literacy-specific listings claims the opposite — postings up roughly 13x in a year while median pay on them slipped about 4%. We couldn’t confirm that second figure against its original source, which is itself the point: treat any single ‘AI skills pay X%’ number, including the ones in this issue, as directional. Explore → | Reddit’s own CEO is betting against the ‘AI ate entry-level hiring’ narrative. Steve Huffman says Reddit plans to ‘go heavy’ on hiring new college grads specifically because they’re ‘AI native’ — the opposite bet from companies quietly freezing junior hiring because AI can already do the work. Entry-level hiring looks like a genuine split right now, not a settled trend in either direction. Explore → |
|
|
From the community Recruiter and hiring-forum commentary, not a vendor survey. The split among recruiters mirrors this issue’s data: some argue AI-powered assessments finally make skills-based hiring real, because software can check actual capability instead of using a credential as a proxy. Others counter that most ‘AI skills required, degree optional’ postings are cosmetic — the same recruiters still default to familiar schools and resumes once an applicant pool gets large, which is exactly the pattern the Burning Glass Institute’s research documented at scale. Staffing Industry Analysts: are skills speaking louder than degrees? →
|
|
FOWL prediction 24 By June 30, 2027, ‘list AI skills on your resume’ stops being enough — a majority of large employers running AI-skills hiring will require a structured proof-of-work check before the claim counts.TestGorilla, Codility, and CodeSignal have all shipped AI-specific readiness and fluency assessments in 2026 specifically because self-reported skills are easy to inflate, and the Burning Glass research shows that policy changes without real verification barely move hiring outcomes — employers who actually want their stated preference to be real have a documented reason to add assessment, not just remove a line from a job ad. The honest counter-case: if these assessment vendors run into the same bias and validity concerns that have dogged algorithmic hiring tools generally, adoption could stall well short of a majority, leaving the self-reported tag as the norm for longer than this prediction expects. Review date: June 30, 2027.
|
|
Put it to work 3 things to do this week.| 01 | If your resume or profile currently just lists ‘AI skills’ as a tag, add one concrete, checkable artifact next to it — a specific workflow you automated, a prompt chain you built, an eval you ran — since a self-reported tag alone is exactly what the newer assessment tools are built to double-check. | | 02 | If you’re hiring, decide now what proof you’ll actually ask for before you publish a ‘no degree required, AI skills a must’ posting — per the Burning Glass research, dropping the line without changing how you screen barely changes who you hire. | | 03 | If you’re job hunting on the strength of AI skills, check which sector the posting is in before you assume the pay follows — PwC’s data puts the premium at 64% in consumer markets and as low as 12% in government and public-sector roles for the same stated skill. |
|
Career radar AI Skills Assessment DesignerAs ‘AI skills over degree’ becomes a hiring claim companies actually make, someone has to build the tests that make the claim checkable — not just a resume keyword. TestGorilla, Codility, and CodeSignal all shipped AI-specific readiness and fluency assessments in 2026, and each needs people who can design valid tests, defend them against bias audits, and keep them from becoming the same checkbox-with-extra-steps problem Burning Glass’s research found with dropped degree requirements. A place to start. Read TestGorilla’s own AI Readiness and AI Fluency assessment launch end to end — it’s a clear, concrete description of what a checkable ‘AI skills’ claim is supposed to look like, which is the real skill this job tests for. Explore AI careers → |
|
|
The opportunity board Know where to look.General AI-training platforms (Outlier, Mercor, DataAnnotation) are still the fastest door into paid AI work regardless of degree, though pay varies hugely by task: Mercor’s own numbers put its average contributor pay around $95/hour versus roughly $30/hour at larger rivals, and project flow isn’t guaranteed — Mercor cut thousands of contractors off a single Meta project in late 2025. Mercor: current AI training projects → On the assessment side, the vendors building these ‘prove it, don’t just claim it’ tools are hiring too — TestGorilla’s own careers page lists roles building and validating the AI-readiness tests this issue covers, no AI-industry background required. TestGorilla: open roles →
|
|
Both halves of this issue’s claim are true at once, which is the uncomfortable part: employers really are removing the degree line, and that really hasn’t changed much about who gets hired. The useful move isn’t picking a side of that contradiction — it’s noticing that a posting’s words and a company’s actual hiring data are two different documents, and only one of them is public by default. Until next Monday,
TYB
Data scientist & founder, FOWL AI
|
|
One question for you Have you ever gotten (or lost) a job specifically because of AI skills, with or without a degree?Hit reply. I read every response, and it helps shape what we cover next.
|
|
Online · Vibe Code Saturdays Read about the future.
Then build it.A free, hands-on Saturday to build a website, newsletter, portfolio, or useful tool with AI. Beginners welcome. Live on Zoom. Next date coming soon — apply and we’ll email your Zoom link and setup videos once it’s set.
|
Know someone who’d find this useful?Share this briefing → Forwarded this email? Get the free weekly briefing.
|
|
|
|