Data current as of August 20, 2026. Numbers on this page snapshot the JobShift dataset on the date above; the AI layoffs view refreshes continuously as new WARN filings land.
Three clusters, half the disclosed worker impact
Three AI-attributed layoff clusters in 2026 account for more than half of the disclosed worker impact recorded in the JobShift dataset: Amazon (~12,100 workers, first effective January 1), Meta (~8,640 workers, first effective March 30 — the largest of three Meta AI-attributed clusters in the 2026 dataset), and Microsoft (~6,105 workers, first effective July 1). Together the three represent 26,845 workers, roughly 54 percent of the disclosed AI-attributed layoff impact in the dataset for 2026 year to date.
The dataset records 31 AI-attributed layoff clusters for 2026 year to date. Nineteen of them (roughly 61 percent) affected fewer than 500 workers or had no disclosed worker count on their WARN filings.
The 19 small or undisclosed clusters
| Company | Workers | States | First effective | |---|---:|---|---| | Uber | 41 | CA | 2026-07-22 | | Clari Inc. | 76 | CA | 2026-02-10 | | Pinterest | 118 | CA | 2026-01-27 | | Vertex, Inc. | 170 | — | 2026-04-28 | | SentinelOne | 230 | — | 2026-05-28 | | Arctic Wolf | 250 | — | 2026-05-06 | | ServiceNow | 287 | CA | 2026-07-28 | | Atlassian US | 315 | CA, WA | 2026-03-11 | | Angi Inc. | 350 | — | 2026-01-07 | | Robinhood | 353 | CA | 2026-06-16 | | Groupon, Inc. | 400 | DE | 2026-05-26 | | Snap | 415 | CA, WA | 2026-04-15 | | Cars.com | not disclosed | — | 2026-04-09 | | DraftKings | not disclosed | — | 2026-02-24 | | Freshworks | not disclosed | — | 2026-08-04 | | GitLab | not disclosed | — | 2026-05-11 | | PEGASYSTEMS INC | not disclosed | — | 2026-01-12 | | Productboard | not disclosed | — | 2026-04-15 | | Snowflake | not disclosed | — | 2026-03-19 |
A dash in the States column indicates the WARN filing did not name a specific state (either coded as generic "US" or left blank in the source data). The twelve disclosed clusters account for 3,005 workers combined. The seven undisclosed clusters could add anywhere from a few dozen to several thousand more; WARN filings that omit worker counts are common in states with looser reporting requirements and are a known limitation of the dataset.
Eight companies on this list — Pinterest, GitLab, Snowflake, ServiceNow, Robinhood, Snap, Uber, and Atlassian US — also had AI-role hiring activity in 2026 and appear in the thirteen-firm overlap set analyzed in the AI-hiring-and-layoffs overlap article.
Sector patterns in the long tail
The nineteen small or undisclosed clusters fall into a handful of recognizable business categories.
Enterprise SaaS and developer tools: Snowflake, GitLab, Atlassian US, ServiceNow, Freshworks, PEGASYSTEMS, Productboard, Vertex, Clari — nine of the nineteen clusters, the largest sector concentration in this set.
Cybersecurity: SentinelOne and Arctic Wolf, both with cluster first-effective dates in May 2026, within three weeks of each other. The JobShift dataset cannot confirm a category-wide pattern from two clusters, but the timing coincidence is on record.
Consumer marketplaces and gig platforms: Cars.com, Groupon, Angi, Uber.
Consumer finance and gambling: Robinhood, DraftKings.
Consumer social and content: Pinterest, Snap.
Between the three mega-clusters and the nineteen small or undisclosed clusters, nine additional AI-attributed clusters ranged from 500 to 4,999 workers each in 2026: Block, Cisco, Cloudflare, Coinbase, Epic Games, Intuit, Oracle, and two additional Meta clusters at approximately 4,665 and 705 workers. That accounts for all thirty-one AI-attributed layoff clusters in the 2026 dataset year to date: three mega-clusters, nine mid-sized clusters, and nineteen small or undisclosed clusters.
What the dataset does and does not confirm
The AI-attributed layoff data captures WARN filings where corroborating press coverage described AI adoption, automation, or AI-driven restructuring as a factor. Several inferences that might follow from the concentration pattern are not supported by the dataset.
Coverage volume is not measured here. The claim that mainstream media coverage focused on the mega-events at Amazon, Meta, and Microsoft is intuitive and probably correct in the aggregate, but the JobShift dataset does not track article counts, headline volumes, or search interest. Readers looking to test that claim rigorously would need to pull GDELT, Google Trends, or Media Cloud data separately.
"Undisclosed" does not mean "small." Seven of the nineteen clusters have no worker count on file. That reflects the WARN filing itself, not JobShift's classification. Some of those seven may be smaller layoffs at smaller firms; others may be large restructurings at firms whose filings omitted the worker count. The JobShift classifier deep-dive at the methodology article documents the full attribution logic.
Small-cluster prevalence is not evidence of a labor-market pattern. The count of 19 small or undisclosed clusters is bounded by WARN reporting thresholds. WARN Act generally requires filings for layoffs of 50 or more workers at a single site; layoffs below that threshold do not appear in the dataset at all. The "long tail" measured here is truncated on the small end, and the true count of small-scale AI-attributed layoffs is unknown.
What the pattern does show
Nineteen of the thirty-one AI-attributed layoff clusters recorded in 2026 to date affected fewer than 500 workers or reported no worker count. The disclosed portion of that group affected 3,005 workers combined, spread across cybersecurity, enterprise software, marketplaces, consumer finance and gambling, and consumer social firms, with named filings in California, Washington, and Delaware. The three mega-clusters at Amazon, Meta, and Microsoft accounted for the majority of worker impact across only three of thirty-one distinct restructuring incidents.
The public narrative of "AI layoffs" is often illustrated with the mega-cluster examples. The 2026 AI-attributed layoff dataset includes twenty-eight additional clusters beyond those three.
Live per-company data
The cluster-level snapshot on this page reflects the JobShift dataset on August 20, 2026. For real-time layoff and hiring activity as new WARN filings and job postings land:
- The AI layoffs view shows AI-attributed layoffs with monthly totals and per-company detail.
- The AI hiring view shows current AI-role postings with company and state breakdowns.
- The WARN filings guide explains how state WARN portals report layoff data and what filing conventions to expect.
- The methodology page documents the AI-attribution logic applied to layoff and hiring events.