Global Tech Layoffs Reach 154,000 in First Half of 2026

It is the pattern: AI restructuring, cost control, and ordinary corporate pruning now arrive in the same envelope, making it harder for enterprise buyers to know whether a vendor is getting sharper or simply hollowing itself out.

The AI story is useful. Too useful.

Gulf News reports that at least 153,965 jobs had been eliminated across the global tech sector as of July 2. That puts the industry on a pace that could challenge the 246,000 layoffs recorded across all of 2025, if the current direction holds.

The named cuts are not coming from fringe operators. Oracle is listed with 25,754 jobs cut so far this year, including a recent round affecting around 600 employees in Romania. Amazon is second after cutting about 16,000 corporate roles in January, following an earlier phase of downsizing that began with 14,000 cuts in October 2025. Cognizant is cutting up to 15,000 roles globally under Project Leap, tied to a $230 million to $320 million cost program. Meta has cut around 10,400 roles across several rounds. Microsoft has reportedly cut about 5,500 roles, around 2.5 percent of its 220,000-person global workforce, though the company has not confirmed an exact number.

The lazy read is that AI is simply replacing workers. The more useful read is colder. AI has become both an operating model and a cover story. Some companies are actually repricing work around automation. Others are doing familiar fiscal trimming and dressing it in language investors prefer.

For enterprise customers, that distinction matters. A vendor that automates delivery may improve margins without breaking service. A vendor that removes experienced staff and calls it “AI-supported” may be creating a new attack vector: thinner support, slower incident response, weaker implementation discipline, and more room for lateral movement when something fails.

The risk moves into delivery, support, and security

The cuts hit different parts of the machine. Microsoft’s reported reductions span sales, consulting, and Xbox. Meta’s rounds touched Reality Labs and several divisions, with a larger May reduction reported alongside cancelled hiring plans. Cognizant’s restructuring is tied to a leaner, AI-supported delivery model, with India expected to take the largest share of reductions because more than 250,000 of its 350,000-plus employees are based there.

That is not just labor-market trivia. Enterprise IT depends on people who do boring, critical work: implementation, migration, patch sequencing, account escalation, audit evidence, managed service handoffs. These functions do not generate keynote slides. They keep negligence from becoming breach material.

Amazon’s case also shows why the label matters. Gulf News reports that Amazon said its WARN notice for around 600 job cuts at a Homestead, Florida logistics facility is tied to converting the two-year-old warehouse from a shipping center into a full-scale fulfillment center, with reopening planned for mid-to-late 2028 and around 1,000 jobs expected to return once work is complete. That is restructuring, not necessarily an AI displacement story.

The same caution applies across the sector. Nearly every major tech company is investing heavily in AI. That does not mean every layoff is caused by AI. It means AI now sits in the same budget conversation as management simplification, cost control, and changing business priorities. Conveniently vague. Operationally important.

What buyers should check before signing

This is where procurement needs less theatre and more forensics. If a supplier has announced cuts, buyers should ask which functions were affected, whether support tiers changed, and whether named technical contacts still exist. Not in a webinar. In writing.

For managed services, consulting, cloud migration, security tooling, and enterprise software, the key question is not whether the vendor “uses AI.” It is whether the vendor can still meet response times, maintain change control, document exceptions, and staff escalations when automation gets something wrong.

There is a capital-market angle too. Durham Post notes rising investor interest in AI and advanced technology companies heading toward future IPO markets, including software platforms, semiconductor developers, cloud infrastructure providers, cybersecurity, robotics, and machine learning firms. That means the same market rewarding AI growth narratives is also listening closely when incumbents describe workforce cuts as efficiency.

That incentive structure is not a conspiracy. It is just capitalism with better slide decks.

The practical takeaway is grim but simple: treat layoffs as a vendor-risk signal, not background noise. Ask what changed inside the delivery chain. Ask what was automated, what was removed, and who owns the failure path. If the answer is polished but unspecific, assume the control has been weakened until proven otherwise.