For two years, the accepted wisdom in tech careers has been simple: artificial intelligence is coming for routine engineering work, so the safe move is to reposition yourself as an AI or machine learning professional. New data out of India suggests that advice may have aged badly — and that the fear has quietly relocated to the very people building the technology.

The survey: AI workers feel as exposed as the roles they were meant to replace

A July 2026 survey of 1,552 India-based professionals by Blind, the anonymous professional networking app, found that 66% of AI/ML workers expect a layoff or significant headcount cut in their team within the next three to six months — one of the highest at-risk readings of any function in the survey. That places AI/ML workers almost level with Sales/Marketing (68%) and Product/Design (65%), the functions that have long been considered the most exposed to cost-cutting.

Engineering, by contrast, comes out looking comparatively secure. Only 24% of engineers rate a cut as “very likely” — the lowest of any function — and overall at-risk sentiment among engineers (58%) sits below the survey average. Data/Analytics professionals feel safer still, at 51% at-risk, with four in ten calling a cut unlikely, the highest confidence score of any team.

In other words, the two job families AI was supposed to hollow out — traditional software engineering and data work — currently feel more secure than the AI/ML function itself.

The unease isn’t rooted mainly in formal layoff announcements, which account for only 9% of the signals at-risk workers cite. It’s the quieter, upstream indicators: hiring freezes (27%) and budget or headcount-target cuts (24%) are doing most of the work in shaping how insecure people feel, long before any layoff is ever announced.

Company-level results sharpen the picture further. Workers at Salesforce (81%) and Oracle (80%) report the highest at-risk sentiment, followed by Uber (77%) and PayPal (75%) — enterprise software and payments firms that have all been mid-restructuring through 2026. At the other end, two Indian-origin SaaS companies stand out for comparatively low anxiety: Zoho at just 9% and Freshworks at 44%.

One Coinbase employee, quoted in a related Blind discussion on engineers’ prospects in India, put the mood bluntly: “ML is honestly dead too… The best you can do is augment your current skills for backend AI pipelines. And ride it out as long as you can.” An Avalara employee went further, arguing that “the only safe ML guys are the ones with a PhD at DeepMind” — a sentiment that captures how narrow the perceived safety zone within AI/ML has become.

The global backdrop: 2026 is already the worst year on record for tech layoffs

The Blind numbers land against a global context that makes the anxiety easy to understand. Tracking firm layoffs.fyi recorded more than 124,000 tech layoffs in the first seven months of 2026, already surpassing the roughly 122,000 cut across the whole of 2025. Other trackers put the number even higher: data compiled by TradingPlatforms and cited by Gulf News showed nearly 154,000 job cuts across the global tech sector by early July 2026.

What distinguishes this wave from earlier downturns is that it isn’t primarily a response to weak earnings. Many of the companies doing the cutting are simultaneously reporting record profits — the layoffs are a deliberate reallocation of capital toward AI infrastructure and AI-capable headcount, not a survival measure.

Some of the defining events of the year:

• Oracle eliminated an estimated 20,000–30,000 roles in one of the single largest cuts of 2026, hitting its cloud and consulting divisions hard, according to reporting compiled by Tech Insider and Gulf News.
• Amazon cut more than 16,000 corporate positions in January 2026, on top of 14,000 cut in October 2025.
• Meta has cut roughly 10,400 roles across multiple rounds this year — including a January reduction in Reality Labs, cuts across five divisions in March, and around 8,000 layoffs in May alongside the cancellation of plans to hire 6,000 new staff.
• Microsoft cut close to 5,500 roles (about 2.5% of its global workforce), concentrated in sales, consulting and Xbox — following an earlier 2026 round of roughly 4,800 cuts, mostly in gaming, according to the Financial Times.
• Block, under Jack Dorsey, eliminated roughly 4,000 jobs — about 40% of its global workforce — in March 2026, explicitly citing “the growing capability of AI tools to perform a wider range of tasks.” It’s regarded as the single largest AI-attributed layoff event in tech history to date.
• IBM’s cuts since September 2024 have topped 15,000, even as the company says it plans to triple U.S. entry-level hiring for AI and hybrid-cloud roles — a pattern of simultaneous cutting and rehiring that shows up across several large employers.
• Cognizant is cutting up to 15,000 roles globally under its “Project Leap” restructuring, and because more than 250,000 of its 350,000-plus employees are based in India, the country is expected to absorb the largest share of the reductions as the company moves to a leaner, AI-supported delivery model.
• Outside the US, Ericsson has cut over 15,600 full-time roles since 2023 (about 15% of its global workforce) amid weaker telecom and 5G spending, and Australia’s WiseTech Global cut roughly 2,000 jobs — about a quarter of its workforce — citing AI automation of supply-chain tasks.

For Indian professionals working abroad, the stakes are compounded by immigration exposure: reporting suggests roughly 12,000 of Oracle’s cuts fell on India-based teams or Indian expatriates, and H-1B visa holders who lose their jobs face a strict 60-day window to secure new sponsorship before they’re required to leave the United States.

India’s IT services sector: contraction and expansion at the same time

Within India, the picture is genuinely two-sided, and headlines tend to capture only one half of it.

On the contraction side, Tata Consultancy Services (TCS) shed 23,460 employees in fiscal year 2026 as part of what it describes as a pivot to an “AI-first” services model — after cutting roughly 12,000 the year before. Wipro cut its fresher hiring target to 7,500–8,000, down from an earlier plan of 10,000, and close to 200 recent recruits have publicly reported onboarding delays stretching past seven months. HCLTech’s stock dropped sharply in April 2026 after cautious full-year guidance. Analysts describe the underlying cause as client companies bringing outsourced work back in-house or replacing it with AI tooling directly — hollowing out precisely the “bench” model that Indian IT services has relied on for two decades, and hitting professionals with three to eight years of experience in delivery-heavy roles hardest.

Quarterly data illustrates how sharp the swing has been: the top four Indian IT firms (TCS, Infosys, HCLTech and Wipro) added a combined 22,622 employees in the June 2025 quarter, but removed a net 9,100 in the same quarter of 2026 — a roughly 31,700-person swing without a comparable change in underlying demand. The damage is concentrated at the entry rung: fewer freshers doing the simple, repetitive first-pass work that AI now handles more cheaply means fewer people getting trained into the senior roles that used to supervise it.

Yet the sector as a whole isn’t shrinking. Industry figures cited by analysts show Indian IT actually added roughly 1.4 lakh (140,000) employees this year, taking total sector headcount to nearly 59 lakh (5.9 million) professionals nationwide — and TCS, Infosys and Cognizant have all signalled plans to hire tens of thousands of freshers in the year ahead. Infosys has kept a target of 20,000 fresher hires for FY26, with 18,000 already onboarded, making it the most active fresher recruiter among the tier-1 firms, though even Infosys is now prioritising candidates with AI and cloud skills over generic engineering profiles.

The more consequential shift may be structural rather than numerical: Global Capability Centres (GCCs) run directly by multinationals — Walmart, JPMorgan, Goldman Sachs, Shell, Caterpillar, Siemens, Apple and others — are hiring aggressively in Bengaluru, Hyderabad and Pune, often at better pay and with less bench-driven precarity than the traditional IT-services model. Analysts tracking the sector describe India’s decades-long era of mass, undifferentiated campus hiring for bulk service delivery as effectively over, replaced by more selective recruitment tied to specific, AI-relevant skills.

Why engineers feel safer than the AI workers replacing them

The apparent contradiction — AI/ML professionals reporting more anxiety than the engineers AI was supposed to displace — becomes more explainable once the layoff data and the compensation data are read together.

First, much of the AI/ML function itself is undergoing internal consolidation. Companies rushed to build AI teams over the past three years; many of those teams are now being resized, re-scoped, or folded into product and engineering groups as the initial hiring wave matures into an efficiency phase. The “safe ML guys,” as the Avalara employee put it, are increasingly seen as a narrow tier — often the PhD-credentialed, foundational-research end of the field — rather than the broader population of ML engineers doing applied or integration work.

Second, engineering hasn’t been spared cuts, but the layoffs skew toward volume-based, deployment-heavy roles — the bench model in Indian IT services, entry-level testing and support work, and generalist delivery roles — rather than core engineering capability. Meanwhile, demand and pay for engineers who can work effectively with AI tools has risen sharply. PwC’s 2026 Global AI Jobs Barometer found a 56–62% wage premium for AI-skilled roles, up from around 25% just a year earlier, after analysing close to a billion job postings — and ManpowerGroup’s 2026 survey of over 39,000 employers found AI skills are now the hardest in the world to hire for, ahead of engineering and IT broadly. The World Economic Forum’s Future of Jobs research projects a net gain of 78 million jobs globally by 2030 (170 million created against 92 million displaced), suggesting the anxiety is less about AI eliminating work overall and more about which specific roles absorb the transition cost first.

That transition cost, on the current evidence, is landing disproportionately on mid-tier AI/ML practitioners, entry-level and bench-based delivery staff, and functions like sales, marketing and product where AI tooling substitutes fairly directly for headcount — while leaving senior, AI-augmented engineers, and workers in Data/Analytics roles that require judgment AI can’t yet replicate, comparatively insulated.

The takeaway

The Blind survey’s central finding — that moving into AI/ML doesn’t make you safer, and may in fact put you in one of the more exposed categories — runs directly against the prevailing career advice of the last two years. Read alongside the global layoff data (Oracle, Amazon, Meta, Microsoft, Block, Cognizant, IBM) and the India-specific numbers (TCS’s 23,460 cuts, Wipro’s fresher slowdown, alongside genuine sector-wide headcount growth and aggressive GCC hiring), a more precise picture emerges: this isn’t AI replacing “AI people” versus “engineers” as job categories. It’s a market repricing specific kinds of work — repetitive, bench-based, entry-level, and now, increasingly, undifferentiated ML/AI integration work — while paying a rising premium for judgment, seniority, and demonstrated fluency in using AI tools rather than simply being labelled an “AI” employee.

For professionals navigating this market, in India or globally, the operative distinction is no longer “which department are you in,” but “how directly does your daily work resemble the kind of routine, high-volume task AI now performs more cheaply” — and how convincingly you can show you’re the person directing that work rather than being replaced by it.

---

Sources

Robert Walters, “The AI Salary premium: Market outlook and strategic implications for the US” (citing PwC 2026 Global AI Jobs Barometer) — https://www.robertwalters.us/insights/hiring-advice/blog/ai-salary-premium.html

Blind, “66% of India’s AI workers brace for headcount cuts, while engineers feel safer,” July 2026 survey of 1,552 India-based professionals

Gulf News, “Tech Layoffs Near 154,000 in 2026 as AI Reshapes Jobs” (July 2026) — https://gulfnews.com/technology/tech-layoffs-near-154000-in-2026-amid-ai-restructuring-1.500595528

Gulf News, “Tech layoffs top 30,000 in 2026. Worst hit countries revealed” — https://gulfnews.com/business/tech-layoffs-top-30000-in-2026-worst-hit-countries-revealed-1.500441093

Notebookcheck, “Tech layoffs in 2026 have already surpassed last year’s total” — https://www.notebookcheck.net/Tech-layoffs-in-2026-have-already-surpassed-last-year-s-total.1357929.0.html

TechCrunch, “Every major tech layoff in 2026 that has name-checked AI” — https://techcrunch.com/2026/07/06/the-running-list-major-tech-layoffs-in-2026-where-employers-cited-ai/

Yahoo Finance / TechCrunch syndication, “Monday.com is the latest tech company to blame AI for layoffs — here are 20 others” — https://finance.yahoo.com/technology/ai/articles/running-list-major-tech-layoffs-012755703.html

Tech Insider, “Tech Layoffs 2026: How AI Is Driving the Biggest Workforce…” — https://tech-insider.org/tech-layoffs-2026-ai-workforce-impact/

InformationWeek, “2026 tech company layoffs” — https://www.informationweek.com/it-staffing-careers/2026-tech-company-layoffs

Outsource Accelerator, “India’s top IT firms are hiring fewer grads” — https://news.outsourceaccelerator.com/india-it-fewer-grads/

LayoffTrends, “IT Layoffs India 2026 — TCS, Infosys, Wipro, GCC Jobs” — https://layofftrends.com/india.html

Sandeep Anand, “TCS, Infosys, Wipro Layoffs 2026: What It Means for You” — https://sandeepanand.in/career-coaching/blogs-tcs-infosys-wipro-layoffs-2026-it-career/

LaPaas Voice, “TCS Layoffs 2026: 23,460 Jobs Gone and What It Means for Indian IT” — https://lapaasvoice.com/tcs-layoffs-2026-indian-it-business-model

Value Add VC, “Future of Work 2026: 92M Jobs Displaced, 170M Created” (citing WEF Future of Jobs Report and PwC 2026 Global AI Jobs Barometer) — https://valueaddvc.com/blog/future-of-work-in-2026-skills-that-ai-cant-replace-and-careers-that-are-safe