For two years, the headline about AI and the workforce was always the same:
AI is coming for your jobs.
Recruiters especially heard it from clients and leadership. “Why hire when AI can do the work?”
Every LinkedIn thought leader with a ChatGPT screenshot had a take.
Today, the data came in. And it says the opposite.
A research paper from Ramp, covered by NBC News, found that companies adopting AI most aggressively are expanding their workforces, not shrinking them. Business Insider’s headline put it bluntly: “So much for the AI jobs apocalypse.”
The U.S. Census Bureau backs this up. Their working paper (CES 26-25, published April 2026) found that AI-related employment decreases happened in only 2% of firms. High-intensity AI adopters increased headcount by 10.2%. Entry-level hiring went up 12%.
So the question for recruiting teams isn’t “will AI replace me?” It’s “am I ready to hire at the volume AI-adopting companies need?”
What the Data Actually Says
The Ramp Study
Ramp, a financial operations company, published research showing that AI’s heaviest adopters are expanding their workforces. NBC News covered it on June 30, 2026. The finding complicates both the “AI will eat all jobs” crowd and the “AI will create utopia” crowd , it depends on how much a company actually invests in AI.
The Census Bureau Numbers
The U.S. Census Bureau’s working paper CES 26-25 used the 2026 AI supplement to the Business Trends and Outlook Survey. During November 2025 to January 2026, 18% of firms used AI in a business function. On an employment-weighted basis, that rose to 32% , meaning larger employers are doing most of the adopting.
Among firms using AI, 57% had integrated it in three or fewer business functions. Sales and Marketing led at 52%, followed by Strategy at 45% and IT at 41%.
The PwC Barometer
PwC’s 2025 Global AI Jobs Barometer found that industries best positioned to adopt AI posted 3 times higher growth in revenue per employee. Wages and job numbers were still growing in virtually every AI-exposed occupation. Workers with AI skills commanded a 56% wage premium.
The skills employers seek are changing 66% faster in the most AI-exposed jobs. That’s reorganization at a pace most recruiting teams aren’t ready for.
Why the “AI Kills Jobs” Story Was Always Lazy
The narrative was convenient. Scary headlines get clicks. “AI will replace 300 million jobs” is a better headline than “AI will reorganize work and probably increase hiring in companies that invest.”
But the story never accounted for what actually happens when a company adopts AI.
They don’t do mistakes like
- Firing people and show savings as SG&A cost cutting.
- They take on more clients (leverage higher output per employee)
- They launch new products. (faster R&D and time to market)
- They enter new markets.
Growth requires people.
PwC’s data shows this clearly , AI-ready industries aren’t just maintaining headcount. They’re growing revenue per employee 3x faster, which means they’re scaling, which means they’re hiring.
The 2% of firms that reported AI-related employment decreases? That’s a rounding error in a labor market of 160 million people. The real story was always about reorganization, not replacement.
But What About the Companies That DID Fire People for AI?
Fair question. saw AI, saw an opportunity to cut headcount, and pulled the trigger. The results were… not great.
Klarna: The Most Expensive “I Told You So” in Fintec
In 2024, Klarna partnered with OpenAI, slashed customer service and marketing departments, and let headcount drop from over 5,500 employees to around 3,500.
The AI chatbot handled 2.3 million conversations in its first month. It was doing the work of 700 customer service agents. Costs fell. Resolution times improved. Every tech newsletter cited Klarna as proof that the AI future was here and it was profitable.
Then reality happened.
Customers started complaining about generic, repetitive responses.
The chatbot could handle simple queries fine. But the moment a situation got complicated, emotional, or unusual , a payment failure because someone was in the hospital, a fraudulent charge during a family crisis , it fell apart.
The chatbot was optimized for closing conversations quickly, not for actually solving problems.
By 2025, Klarna was rehiring human customer service staff.
Dukaan: 90% Gone in One Day
In July 2023, Dukaan announced that they had replaced 90% of his customer support staff with an AI chatbot, and about cutting support costs by 85%. Response times dropped from nearly 2 hours to under 4 minutes.
The backlash was immediate.
Customers complained about wrong answers, broken escalation paths, and a complete inability to handle anything beyond basic FAQs.
This move was met with a negative response for celebrating layoffs while customers suffered.
The Pattern Is Clear
Both companies made the same mistake: they treated AI as a replacement for humans instead of a tool that augments humans.
They optimized for cost and speed.
They forgot about quality, edge cases, and the fact that customers are human beings who sometimes need a human to talk to.
What This Means for Recruiting Teams
More Requisitions, Same Team
If your company is adopting AI (and most are), you’re probably looking at more open roles, not fewer. The Census data shows adoption concentrated in large firms and knowledge-intensive sectors , exactly the companies that hire the most.
Your recruiting team size hasn’t changed. Your req load just went up.
Skills Are Shifting Faster Than Job Descriptions
PwC found that skills sought by employers are changing 66% faster in AI-exposed jobs. That means the candidate profile you hired for six months ago may not match what you need today.
Job descriptions are stale the moment they’re written. Screening against outdated criteria = you filter out the people you actually need.
Entry-Level Volume Is Up
The Census data shows entry-level headcount rose 12% in high-intensity AI adopters. That’s high-volume hiring. Entry-level roles generate the most applications per req. If you’re screening 400 applicants per entry-level role manually, and you now have 30% more entry-level reqs, you’re drowning.
The Wage Premium Problem
Workers with AI skills command a 56% wage premium. That means the candidates you want are more expensive and harder to close. Sourcing matters more. Assessment matters more. You can’t afford to waste time on candidates who look good on paper but can’t actually do the work.
The Irony Nobody’s Talking About
AI was supposed to reduce the need for hiring. Instead, it’s increasing it.
Companies adopt AI → they grow faster → they need more people → recruiting teams get crushed under higher volume → they need AI to handle the volume.
The same dynamic played out with every major technology shift. ERP systems didn’t eliminate accounting departments. They made companies bigger, which meant more accounting work.
The difference now is speed. AI adoption is happening faster than any previous technology shift. The Census data projects adoption will reach 22% of firms within six months. Recruiting teams that don’t adapt their process will drown in volume.
What We’ve Learned Working with Recruiting Teams
I’ll be upfront , I’m biased here because we built Gappeo. But the patterns we’ve seen across mid-market and high-volume hiring teams are consistent regardless of what tool you use.
Volume Is the Killer, Not Quality
Almost every team we work with can identify good candidates when they have time to look. The problem is they don’t have time. When req load doubles and your team stays the same size, quality drops because recruiters start skimming. They don’t have time not to.
Screening Is Where Time Goes
Phone screens eat 60-70% of recruiter time on high-volume roles. That’s the bottleneck. If you can automate first-round screening , not with a keyword filter, but with something that actually evaluates candidates against job criteria , you buy back the hours your team needs to focus on closing.
Consolidation Beats Stacking
The teams that handle volume best aren’t the ones with the most tools. They’re the ones with the fewest. Five tools means five logins, five data silos, and five places where candidate information gets lost. One platform that handles screening, assessment, and engagement = your recruiters work from one dashboard instead of stitching data together across five tabs.
If you’re dealing with the hiring volume that AI adoption creates, automating your phone screening is the fastest way to buy back recruiter hours. And if you’re weighing whether AI phone screening actually works compared to human screening, the answer is: for first-round filtering at volume, it works better. Humans are better for final rounds and relationship-building.
Speed to Value Determines Whether You Survive the Volume Spike
Teams that implement tools in days handle the volume spike. Teams that take 8 weeks to implement get buried before the tool is live. When your req load suddenly increases because your company just adopted AI across three departments, you don’t have two months to wait.
What to Actually Do Right Now
If you’re a recruiting leader reading the news today and realizing your hiring volume is about to go up (or already has), here’s the practical version:
Audit your current stack. Count how many tools your recruiters log into daily. If it’s more than two for screening and assessment, you’re paying for fragmentation.
Identify your bottleneck. Is it screening volume? Scheduling? Assessment? Pick the one eating the most hours and automate that first. Recruitment workflow automation gives your team capacity to handle more reqs without burning out.
Test with real candidates. Don’t evaluate tools in demo environments. Run 50 real candidates through during your trial. If the tool can’t handle your actual workflow with your actual candidates, the demo doesn’t matter.
Look at the best recruitment automation tools for staffing agencies and internal teams. The landscape is shifting fast. What worked last year may not handle the volume AI adoption is creating this year.
If you want to see what consolidated AI screening and assessment looks like with your candidates, book a demo. Or start a free trial , no credit card, no 8-week implementation.
Frequently Asked Questions
Is AI actually creating jobs or just changing them?
Both. The Census Bureau data shows AI-related employment decreases in only 2% of firms. High-intensity adopters increased headcount 10.2%. The jobs are changing , skills sought are shifting 66% faster in AI-exposed roles , but the net effect is more hiring, not less.
What does the Ramp study actually say?
Ramp’s research, covered by NBC News on June 30, 2026, found that AI’s heaviest adopters are expanding their workforces. The key variable is how much a company invests in AI , companies that go deep see growth that requires more hiring.
If companies are hiring more because of AI, do recruiters still need AI tools?
More than ever. Higher hiring volume with the same team size = you need automation to keep up. AI adoption creates the exact problem (volume) that AI recruiting tools solve.
What should recruiting teams prioritize right now?
Screening automation. If your company is adopting AI and growing, your req load is going up. Screening is where 60-70% of recruiter time goes on high-volume roles. Automating first-round screening buys back the hours your team needs.
How does Gappeo fit into this?
Gappeo consolidates screening, assessment, and engagement into one platform. If your hiring volume is increasing because your company is growing (partly thanks to AI), you need a tool that handles volume without requiring you to buy four more tools. Pricing starts at $29/month with a free trial.
Works Cited
Barr, Alistair. “So Much for the AI Jobs Apocalypse. Demand for Tech Talent Is Rising.” Business Insider, 25 June 2026, www.businessinsider.com/ai-supposed-to-kill-tech-jobs-instead-demand-up-2026-6. Accessed 30 June 2026.
“Indian CEO of Dukaan Replaced 90% of Staff with an AI Chatbot.” CNN, 12 July 2023, www.cnn.com/2023/07/12/business/dukaan-ceo-layoffs-ai-chatbot. Accessed 30 June 2026.
“Klarna Reverses AI Push, Says Customers Prefer Human Support.” Forbes, 18 May 2025, www.forbes.com/sites/quickerbettertech/2025/05/18/business-tech-news-klarna-reverses-on-ai-says-customers-like-talking-to-people. Accessed 30 June 2026.
“Klarna CEO Admits Aggressive AI Job Cuts Went Too Far, Starts Hiring Again After US IPO.” MLQ News, 2025, mlq.ai/news/klarna-ceo-admits-aggressive-ai-job-cuts-went-too-far-starts-hiring-again-after-us-ipo. Accessed 30 June 2026.
Park, Lisa. “High-Intensity AI Adopters Are Hiring More, Including Entry-Level Workers.” Prism News, 30 June 2026, www.prismnews.com/news/high-intensity-ai-adopters-are-hiring-more-including-entry. Accessed 30 June 2026.
Perlo, Jared. “New Data Finds AI’s Heaviest Adopters Are Expanding, Not Shrinking, Their Workforces.” NBC News, 30 June 2026, www.nbcnews.com/tech/tech-news/ai-jobs-data-study-hiring-economy-rcna352206. Accessed 30 June 2026.
“AI Linked to a Fourfold Increase in Productivity Growth and 56% Wage Premium, While Jobs Grow Even in the Most Easily Automated Roles: PwC Global AI Jobs Barometer.” PwC, June 2025, www.pwc.com/gx/en/news-room/press-releases/2025/ai-linked-to-a-fourfold-increase-in-productivity-growth.html. Accessed 30 June 2026.
Ramp. “Ramp AI Index.” Ramp, 2026, ramp.com/data/ai-index. Accessed 30 June 2026.
U.S. Census Bureau. “The Microstructure of AI Diffusion: Evidence from Firms, Business Functions, and Worker Tasks.” Center for Economic Studies Working Paper Series, Working Paper No. CES-26-25, Apr. 2026, www.census.gov/library/working-papers/2026/adrm/CES-WP-26-25.html. Accessed 30 June 2026.
Leave a Reply