Two reports landed a day apart at the start of October, and together they explain why so many 2027 headcount plans feel contradictory. On October 1, 2026, outplacement firm Challenger, Gray & Christmas published its September job cuts report. Through September, US-based employers announced 573,195 job cuts, down 39% from the first nine months of 2025.
AI has been cited in 120,136 of those announcements, approximately 21% of all cuts, and it remains the leading reason year to date. Technology companies have announced 165,925 cuts this year, up 54% from the same period in 2025, and technology accounts for 29% of all job cuts announced in 2026, more than any other industry.
On October 2, CompTIA published its analysis of September tech employment and job postings. Active job postings requiring AI-related skills reached 349,821, up 32,327 from August and up roughly 282% since May 2024. Total active tech postings reached 625,633, the highest in more than three years.
In the same month, tech sector employment fell by an estimated 10,350 positions, and the tech unemployment rate edged up to 3.3%, still below the national rate of 4.2%. So which story is true? Both, and the reason matters for your planning.
A disclosure before we go further. Golabs recruits and staffs AI engineers, so rising demand for AI skills is good for our business. Challenger, Gray & Christmas sells outplacement services, and CompTIA sells certifications. We flag each source’s interest where it applies.
Are AI Layoffs and AI Hiring Contradicting Each Other?
No. They describe different seats. Cuts citing AI are about roles employers expect to automate or consolidate, while postings requiring AI skills are for people who can build, use, and run AI. Challenger counts 120,136 AI-attributed cuts so far in 2026, and CompTIA counts 349,821 active postings requiring AI skills. The labor market is changing which skills it pays for.
Indeed Hiring Lab’s September 17, 2026 analysis shows the same shift in job postings. In the occupations most exposed to generative AI, including software development, IT systems and support, and data and analytics, the entry-level share of postings listing an annual salary fell from 29% to 10% between 2021 and 2026, while the senior share rose from 22% to 47%.
Pay moved with it. Advertised pay in the most AI-exposed occupations has risen by about 46% since 2021, compared with about 25% in the least exposed. Controlling for occupation and month, Indeed estimates a post-ChatGPT pay premium of 5.7%.
Controlling for seniority as well, the premium falls to 2.4% and is no longer statistically significant at the 5% level. Indeed notes that some of the rise reflects a shift toward fewer, more senior postings, though this control may over-correct.
Read together, the three sources describe one market reshaping itself. Employers are cutting some roles, posting fewer entry-level openings in AI-exposed fields and paying more for experienced people who can work with AI. That reading is ours, but each piece of it is in the data.
How Much Should You Trust the AI Layoffs Number?
Enough to take seriously, not enough to copy into your plan as is. The headline figure is real, but it measures something narrower than “AI is taking jobs.” Three caveats matter before it shapes your plan.
These are announcements, not completed layoffs. Challenger tracks announced plans by US-based employers. Announced cuts can be delayed, reduced, or spread over years, and the reason an employer gives is the reason it chose to state publicly.
The monthly pace has slowed. In September alone, AI was the fifth most cited reason, with 3,961 cuts, about 9% of the month’s 43,281 total. Andy Challenger summarized the trend this way: “We’ve seen layoff activity subside over this year, and September continues to illustrate this point.”
Hiring is weak too. Employers announced plans to hire 90,787 workers in September, the lowest September total since 2011, though year-to-date hiring plans of 210,612 are up 3% from 2025. The Bureau of Labor Statistics reported on October 2 that nonfarm payroll employment (+29,000) and the unemployment rate (4.2%) changed little in September.
Our reading is a market with low hiring and slowing cuts, where employers are cautious about adding headcount. That is different from a market where AI is replacing workers at scale. For headcount planning, the risk of under-hiring for the right skills is as real as the risk of over-hiring.
What Does the AI Hiring Number Actually Measure?
The 349,821 figure counts active job postings requiring AI-related skills, using Lightcast data. CompTIA notes that BLS and Lightcast data may be subject to backward revisions. It is a measure of how often employers ask for AI skills, not a count of open “AI jobs.”
The more telling number sits next to it. Active postings in CompTIA’s AI and machine learning job titles group totaled 13,636, nearly triple the May 2024 level but small next to the 349,821 postings asking for AI skills. Most AI demand, in other words, is for AI skills inside existing roles, not for new AI titles.
The roles carrying that demand are familiar ones. Software developers and engineers led all tech occupations with 47,598 new openings in September, up 8.3% from August, followed by systems engineers, tech support specialists, data analysts, and DevOps engineers. Sixteen of the 20 industries CompTIA tracks saw month-over-month increases in tech postings.
That shapes how you hire. If you write a 2027 plan around a handful of new “AI engineer” requisitions, you are planning for the smaller share of the demand. The larger share is your existing engineering, data, and operations roles, rewritten to require the ability to work with AI.
We looked at who you compete with for these people and who you are actually competing with for AI engineers. Note that the 86,000 August gain in tech occupation employment we reported there did not repeat: CompTIA estimates tech occupation employment fell by roughly 6,000 in September.
Which Should Drive Your 2027 Headcount Plan?
Neither headline. The work you need done in 2027 should drive the plan, and both numbers are signals about what that work costs and where to find the people for it. Use them to test your assumptions, not to set your targets.
Do not backfill by job title. When a role is vacated or cut, ask what work remains, what agents or tools now handle, and what new work has appeared, before you post the old job description again. The roles employers are cutting and the roles they are posting are rarely the same seat.
Redeploy before you release. People who already know your systems, customers, and data are often the fastest path to AI-capable roles, if you fund the training and the time. Our guide to preparing your workforce for AI adoption covers the rollout, and our AI fluency program is one structured option.
Write AI skills into existing roles. With AI skills appearing in far more postings than AI job titles do, update the job descriptions for your engineers, analysts, and DevOps staff to say what AI work each role does and how you will assess it in interviews.
Budget for seniority. If AI-exposed postings keep tilting senior, the people you most need in 2027 will cost more. We covered 2027 pay ranges for what you should pay AI engineers, and the case for keeping an entry pipeline anyway, should you still hire junior developers in the AI era.
Price the risk of cutting too early. Gartner predicted in September 2026 that by 2029, 30% of employees laid off due to AI-driven replacement will need to be rehired, often at a significantly higher cost. That is a prediction, not a measurement, but it is a reason to cut only where the work has actually moved.
Keep flexible capacity for seats you cannot size yet. For roles where AI’s effect is still unclear, contract or partner capacity lets you learn before committing permanent headcount. Our guide to planning AI skills for 2027 covers when to build, buy, or borrow, and AI talent solutions is how we help.
A caution about our own corner: no independent 2026 dataset on Latin American AI engineering supply or pay matches the US sources above, so ask any nearshore partner, us included, for rates and timelines in writing. Precise savings figures from any vendor describe that vendor’s own book of business.
Check the signals monthly. Challenger and CompTIA both publish every month, and one month of data is not a trend. Put a monthly review of both on the calendar, and adjust hiring plans each quarter rather than once a year.
Common Questions About AI Layoffs and AI Hiring
Is AI really the top reason for layoffs in 2026?
Among announced job cuts, yes. Challenger, Gray & Christmas reports that AI has been cited in 120,136 announced cuts through September 2026, about 21% of the total, making it the leading reason year-to-date. In September alone, AI ranked fifth with 3,961 cuts. These are announced plans, with the reasons employers chose to state.
How many job postings require AI skills?
CompTIA counted 349,821 active job postings requiring AI-related skills in September 2026, up 32,327 from August and up roughly 282% since May 2024. Only 13,636 active postings were in CompTIA’s AI and machine learning job titles group, so most demand is for AI skills inside existing roles.
Should companies cut headcount because of AI in 2027?
Only where the work has actually moved. Cutting ahead of proven change risks rehiring later at a higher cost, a risk Gartner predicts will affect 30% of employees laid off due to AI replacement by 2029. A safer approach is to redeploy people first, rewrite roles around AI skills, and keep flexible capacity for roles you cannot size yet.
Do AI-skilled workers get paid more?
Generally yes, though part of the premium reflects seniority. Indeed Hiring Lab found advertised pay in the most AI-exposed occupations rose about 46% since 2021, versus about 25% in the least exposed. Its estimated 5.7% post-ChatGPT premium falls to 2.4% and is no longer statistically significant after controlling for seniority.
If you are building your 2027 headcount plan and want a second opinion on which roles to redeploy, rewrite, or fill, talk to the Golabs team. We will start with the work, and we will tell you where hiring is not the answer.