Signals & Shifts: What 'Stable' Is Hiding

This morning the Bureau of Labor Statistics released the March 2026 JOLTS report. Job openings at 6.9 million, largely unchanged from February. Layoff rate at 1.2%. Headline read: stable.

“Stable” is doing a lot of work in that sentence.

Signal 1: The information sector is fracturing — and the headline won’t tell you

#

The national layoff rate ticked up from 1.0% to 1.2%. The information sector’s layoff rate moved from 1.3% to 2.4% — nearly double — in the same twelve months. That is the largest increase of any sector tracked in March’s JOLTS release, and nearly five times the national average increase (Indeed Hiring Lab, May 5, 2026).

Professional and business services followed, climbing 0.7 percentage points year-over-year. Government job openings are down 13% year-over-year, with federal openings down 41% since March 2025 — a direct product of federal workforce reduction initiatives. Information sector job openings are down 33% year-over-year, the steepest decline of any private sector tracked by Indeed’s real-time postings index. IT infrastructure, operations, and support roles are running approximately 30% below their pre-pandemic levels.

The JOLTS narrative also contains a forward signal: recent layoff announcements in technology have explicitly cited AI as the driver, with firms explaining that AI-enabled automation is absorbing tasks that previously required dedicated headcount. Those cuts are still working through the reporting system — as Indeed Hiring Lab noted, “March may be foreshadowing what April and May will confirm.”

Contrast the information sector’s trajectory with retail trade (job openings up 58% year-over-year) and manufacturing (up 18%). Healthcare and education continue their structural growth. These sectors are not experiencing the same labor market. A software engineer and a retail operations manager looking at the same national headline are reading entirely different documents.

For professionals in information, technology-adjacent services, or professional and business services: the headline number is not your number. The fault line in your sector is already visible in the data and is moving.

Signal 2: The entry-level floor has collapsed

#

The second fracture runs between experience levels, and it opened faster than most labor market analysts anticipated.

The unemployment rate for recent college graduates hit 5.7% in Q4 2025 — the highest in three years, and meaningfully above the national average for all workers (Indeed Hiring Lab, April 23, 2026). Since mid-2023, new labor market entrants have accounted for 85% of the total rise in unemployment, according to Oxford Economics. The share of unemployed Americans who are new workforce entrants has reached its highest level in 37 years.

The hiring-side numbers are equally blunt. Job postings targeting new graduates on Handshake, the campus-focused platform, fell 15–16% year-over-year between August 2024 and August 2025, while applications per posting rose 26–30%. Only 30% of 2025 graduates secured full-time employment in their field, down from 41% the year before. The job-finding rate for college graduates — the share of unemployed grads who find work in a given month — is now nearly equal to the job-finding rate for high school graduates, a convergence that would have been difficult to predict a decade ago.

On Indeed, junior-level postings fell 7% year-over-year in 2025 while senior-level postings rose 4%. Internship postings, which serve as the primary pipeline into full-time entry roles, ended 2025 at their lowest level since 2020. The combination of AI automating entry-level task work, employer caution in a slow-hire environment, and explicit seniority requirements has effectively redrawn the bottom rungs of the career ladder.

The entry-level floor has not disappeared entirely. It has been rebuilt with different materials: fewer structured programs, more explicit output requirements, higher minimum evidence thresholds. “Ready to contribute” is no longer sufficient positioning. “Here is what I have already done and measured” is now the table-stakes entry.

Signal 3: AI skill demand is outrunning AI skill supply — and the gap is wide open

#

The third signal moves in a different direction from the first two, and it represents the most immediately actionable opening in the current market.

Job seeker searches for AI-related roles on Indeed have grown 11-fold since the launch of ChatGPT in November 2022 (Indeed Hiring Lab, April 28, 2026). That sounds like a saturated market for AI talent. It is not.

Despite 11x growth in searches, AI-related job searches still represent less than 1% of all searches on Indeed in early 2026. The Indeed AI Tracker simultaneously shows that nearly 5% of all job postings now include AI or adjacent skill requirements. The ratio: employers looking for AI-capable candidates are running five times ahead of job seekers actively positioning for those roles.

This is not a narrow gap in a niche function. The AI Tracker shows AI skill requirements spread across 45% of data and analytics postings, 15% of marketing postings, and 9% of HR postings. The demand has moved well past the engineering org and into operational functions. Yet the supply signal — professionals explicitly positioning themselves as AI-capable and searching for AI-linked roles — has not kept pace.

Jensen Huang made the counterintuitive argument at the Milken Institute on May 4 that AI’s automation of individual tasks does not automatically eliminate the broader purpose of a role (TechCrunch, May 4, 2026). That framing has limits — the tech sector layoff data above complicates any simple “AI creates as many jobs as it eliminates” argument. But his underlying point about task-versus-role has operational value: professionals who can position themselves as the human judgment layer operating above AI tooling are occupying the function employers are actively paying for.

The 5:1 demand-to-search ratio is a visible, measurable skills gap with a closing window. As AI tool adoption continues to accelerate across every function, more professionals will eventually signal AI capability. The gap will narrow. The professionals who document and position that capability now have a compounding advantage over those waiting for the signal to be clearer.

The shift to make this week

#

Three fault lines, three concrete responses:

1) Locate yourself on the sector map

#

Pull up job postings data for your specific function within your specific industry. Is your sector in the information or professional-services cluster where layoffs are rising and openings have contracted 33%? Are you in retail, manufacturing, or healthcare where the dynamics are structurally different? Determine whether your personal labor market resembles the 1.2% national layoff rate — or the 2.4% rate in the information sector. Act according to your actual map, not the national summary.

2) If you are early-career, compete on demonstrated output

#

The 2026 entry-level market does not reward credential accumulation or expressed potential. Employers are operating in a market with more applicants per posting than at any point in several years, and they are increasingly requiring AI-adjacent task capability from new hires who previously would have learned on the job. One concrete, quantified piece of delivered work — a process you accelerated, a cost you identified, a decision you structured — outperforms six months of additional education in the current hiring signal hierarchy. Reframe every application from “ready to contribute” to “here is what I have already done.”

3) Step into the AI skills gap while it remains wide

#

The 5:1 employer-demand-to-job-seeker-search ratio is a structural window that will not stay open indefinitely. Identify one AI-adjacent capability explicitly required in live job postings in your function — not a general “AI knowledge” gesture, but a specific tool, workflow, or output type. Build visible evidence of it this month. Publish or document the result. The professionals closing this gap now are building the kind of compounding advantage that does not show up immediately in a salary offer but becomes decisive when the gap narrows in twelve to eighteen months.

The March 2026 JOLTS number is technically accurate. 6.9 million openings, 1.2% layoff rate, 3.5% hires rate. The market is, in aggregate, stable. But aggregate stability is not personal stability. Three fault lines are widening beneath that number simultaneously: one fracturing the information sector from the broader economy, one separating experienced workers from new entrants, and one separating AI-capable professionals from everyone else. The national headline tells you where the average is. It does not tell you which side of each fault line you are standing on — and the time to look is before the gap becomes uncrossable.

References

#

• Indeed Hiring Lab. (May 5, 2026). “March 2026 JOLTS Report: Stable, Depending on What You Do.” https://www.hiringlab.org/2026/05/05/march-2026-jolts-report-stable-depending-on-what-you-do/ (Accessed May 5, 2026)
• Indeed Hiring Lab. (April 23, 2026). “New Grads Looking for Work: The Struggle Is Real.” https://www.hiringlab.org/2026/04/23/new-grads-looking-for-work-the-struggle-is-real/ (Accessed May 5, 2026)
• Indeed Hiring Lab. (April 28, 2026). “Job Seeker Searches for AI Roles Have Grown.” https://www.hiringlab.org/2026/04/28/job-seeker-searches-for-ai-roles-have-grown/ (Accessed May 5, 2026)
• Ropek, Lucas. (May 4, 2026). “As Workers Worry About AI, Nvidia’s Jensen Huang Says AI Is Creating an Enormous Number of Jobs.” TechCrunch. https://techcrunch.com/2026/05/04/as-workers-worry-about-ai-nvidias-jensen-huang-says-ai-is-creating-an-enormous-number-of-jobs/ (Accessed May 5, 2026)