Two new reports dropped this week that paint starkly different pictures of where the AI job market is headed. One says 93% of jobs face AI disruption six years ahead of schedule. The other shows AI isn’t causing mass unemployment yet, but is quietly strangling the entry-level hiring pipeline.
Meanwhile, IBM is doing something almost nobody else will: tripling entry-level hiring while warning that companies cutting junior roles are hollowing out their future leadership.
The $4.5 Trillion Shift
Cognizant’s “New Work, New World 2026” report revised its forecast of AI’s impact sharply upward. The consulting firm now estimates AI could handle $4.5 trillion in U.S. work tasks and affect 93% of jobs.
That’s not a typo. Ninety-three percent.
The headline number is alarming, but the detail that should concern workers most is the timeline. Cognizant originally projected this level of disruption wouldn’t arrive until 2032. “What we projected might take until 2032 to unfold is happening now before our eyes,” the report states.
The firm also raised its estimate of jobs facing existential threat from AI to 30%, up 15 percentage points from its initial assessment. Professional services, data entry, basic coding, customer support, and content creation are seeing the fastest displacement.
Anthropic’s Data: It’s Not Layoffs, It’s Hiring
Anthropic released its own analysis this month, titled “Labor Market Impacts of AI.” The study takes a different approach: rather than predicting what AI could theoretically do, it measures which tasks AI is already handling based on real Claude usage patterns.
The findings are more nuanced than Cognizant’s projections. There’s no unemployment crisis to report yet. The study found no measurable rise in joblessness among workers in high-exposure occupations since ChatGPT launched.
But the crack is showing up in hiring instead. Among workers aged 22 to 25, the monthly job-finding rate in high-exposure occupations has fallen roughly 14% since ChatGPT’s arrival.
Computer programmers, customer service representatives, and financial analysts are among the most exposed roles. These aren’t jobs being eliminated en masse. They’re jobs that are getting harder to break into.
The Entry-Level Collapse
This aligns with broader data showing entry-level tech job postings down 73% year over year, while AI-related roles grew 163%.
Entry-level hiring at the 15 biggest tech firms fell 25% from 2023 to 2024, and the decline has continued into 2026. The reason is structural: entry-level positions exist primarily for training and scaling teams. When companies shrink, they cut the roles requiring the most mentorship.
AI makes this worse because it’s particularly good at the tasks traditionally assigned to juniors: writing boilerplate code, generating standard functions, debugging syntax errors, formatting documents, and handling routine customer inquiries. The work that companies used to pay beginners to learn while doing is increasingly handled by Claude or GPT.
Job postings now routinely demand two to three years of experience for what used to be entry-level roles. The cruel paradox: you need the job to get the experience, but you need the experience to get the job.
IBM’s Contrarian Bet
One major tech company is swimming against this current. IBM announced in February it would triple entry-level hiring in 2026.
The reasoning is counterintuitive. IBM’s chief human resources officer argued that “the companies three to five years from now that are going to be the most successful are those companies that doubled down on entry-level hiring in this environment.”
The logic: if everyone cuts junior roles today, there won’t be mid-level talent tomorrow. Companies will face a leadership pipeline crisis when today’s seniors retire or leave.
But IBM isn’t just hiring for the same old roles. They’ve redesigned entry-level jobs to focus less on tasks AI can automate and more on customer engagement, problem solving, and collaboration. Junior developers at IBM now spend less time writing routine code and more time working directly with clients.
It’s an experiment worth watching. If IBM is right, the companies slashing entry-level positions are setting themselves up for long-term talent problems. If they’re wrong, they’re paying for training that AI could handle for free.
The Salary Split
For those who do break in, the rewards are substantial. AI engineer salaries averaged $206,000 in 2025, a $50,000 jump from the previous year. Specialists in deep learning, LLM fine-tuning, and MLOps command premiums of 30-50% above generalist engineers.
Entry-level AI roles start between $120,000 and $150,000. Mid-career engineers with three to five years of experience land between $150,000 and $220,000. Senior roles clear $300,000 with equity and bonuses.
But the denominator is shrinking. Despite premium pay, companies report struggling to fill positions. Job postings for AI talent have nearly doubled year over year, but fewer pathways exist for workers to develop those skills on the job.
The skills gap is becoming a moat. Domain experts command salaries 30-50% higher than generalists with equivalent experience. Over 75% of AI job listings specifically seek deep, focused expertise.
What This Means
The job market is splitting in two.
For experienced AI practitioners with specialized skills, demand is strong, compensation is exceptional, and job security appears solid. The talent shortage for senior roles is real.
For everyone else, particularly those trying to enter the field, the path is narrowing. Entry-level roles are disappearing. The traditional career ladder that trained junior talent into senior talent is breaking down. Companies want people with experience they can’t get without jobs that don’t exist.
Cognizant and Anthropic are measuring different things but reaching compatible conclusions. AI isn’t causing mass unemployment because workers aren’t losing jobs they already have. But it is reshaping who gets hired in the first place.
The invisible layoff, as one CEO called it, isn’t about firing people. It’s about never hiring them to begin with.
What You Can Do
If you’re trying to break into AI-adjacent tech work:
-
Build with AI, not around it. Demonstrating comfort with AI tools is becoming table stakes. 13.3% of entry-level postings now explicitly require AI skills.
-
Go deep, not broad. Generalists are competing with AI for basic tasks. Specialists in narrow domains command better pay and have more job security.
-
Target contrarian companies. IBM isn’t alone in betting on human talent. Companies prioritizing training over automation may offer better entry points.
-
Document everything. Without traditional entry-level work history, portfolios of projects, contributions to open source, and demonstrated problem-solving matter more.
-
Consider adjacent roles. Customer-facing positions, project management, and roles requiring human judgment may be safer bets than pure technical roles that compete directly with AI capabilities.
The job market of 2032 arrived in 2026. The companies that figure out how to develop talent rather than just consume it will have an advantage. For workers, the path is harder than it was five years ago, but it’s not closed. It’s just different.