The AI Era Is Bringing Recurring Layoffs

Source: Business Insider

Published: 2026-07-12

Entity Analyzed: Big Tech Operational Workforces

URL SCAN

Tech companies are making recurring layoffs as a form of ‘continuous tuning’ rather than crisis-driven cuts. Mentions of layoffs alongside AI on corporate earnings calls have surged from fewer than five per quarter in 2022 to over 100 per quarter this year.

The Triage

The mask is off, and the language has shifted. What was once called ‘restructuring’ or ‘rightsizing’ is now openly discussed as ‘continuous tuning’—a phrase that carries all the moral weight of calibrating a thermostat. Joseph Fuller at Harvard Business School gives it academic legitimacy. The reality is simpler: tech companies have spent a quarter-century cutting costs, have run out of fat, and are now carving into muscle because the competitive pressure to demonstrate AI productivity gains has made any other path seem like surrender.

The article captures the essential contradiction without naming it: these companies are profitable, growing, and investing heavily in AI—yet they keep cutting workers. Cloudflare cut 20% of its workforce while growing 30%. Microsoft eliminated 4,800 jobs while remaining highly profitable. The cuts are not about financial survival. They are about signaling. Boards are pressuring management teams to show AI-driven productivity gains without increasing spend on tokens or compute. The workers are collateral damage in a demonstration project.

The Autopsy (with DT-LAG)

Mechanical Collapse Point

The collapse point is not the layoffs themselves. It is the institutionalization of ‘continuous tuning’ as a permanent management strategy. When layoffs shift from recession-era crisis response to ‘the way we operate,’ the employment contract is functionally void. The article notes that companies don’t yet know how AI will reshape their businesses—but they are cutting anyway. This is not optimization. It is ritual sacrifice. The cuts happen because the narrative demands them, not because the technology enables them.

Carrol Chang’s observation is the most mechanically significant detail: ‘Truly AI-native and AI-fluent workers are incredibly scarce, and when you can find them, they’re incredibly expensive.’ This means companies are cutting operational workforces without having the AI capability to replace them. The result is not automation—it is under-staffing dressed in AI clothing. The ‘continuous tuning’ produces smaller, more stressed teams doing the same work with fewer resources, while executives report ‘efficiency gains’ to boards.

Lag-Weighted Social Timeline

Social recognition of this pattern will lag 18-24 months. Currently, the framing is neutral-to-sympathetic: ‘continuous tuning’ as adaptation, layoffs as ‘unfortunate but necessary.’ The recognition lag is extended by three factors: (1) the tech industry’s cultural mythology of ‘talent’ and ‘innovation’ makes workers identify with the companies cutting them; (2) the job market for AI-native workers creates a visible success story that masks the broader collapse; and (3) the intermittent nature of the cuts—small waves rather than mass events—prevents the formation of collective response.

Jeffrey Pfeffer at Stanford sees it clearly: repeated layoffs create lasting uncertainty, encourage top performers to leave, and weaken the institutional knowledge that makes companies effective. The rehiring of workers for roles previously eliminated—what the article calls the ‘sneaky truth’ of AI layoffs—is not a recovery. It is proof that the cuts were premature and poorly conceived. But by the time the rehiring happens, the damage to morale, trust, and institutional memory is already done.

Lag Factors

– Stock Option Vesting: Golden handcuffs delay departure decisions. Workers stay through layoff rounds hoping their equity vests before the next cut.
– Cultural Mythology: The ‘talent’ narrative makes workers complicit in their own devaluation. Being laid off by a ‘prestigious’ tech company is still framed as a resume event rather than exploitation.
– AI Skills Theater: Companies invest in ‘AI literacy’ programs that train workers to use tools rather than to build them. This creates the illusion of upskilling while the actual productive work moves to AI-native vendors and elite engineering teams.
– Geographic Concentration: Tech workers in SF, Seattle, and NY have limited relocation options without abandoning their networks and housing investments.
– Rehiring as Evidence: The fact that some companies rehire for eliminated roles is interpreted as ‘the market correcting itself’ rather than ‘management made preventable mistakes.’

Defensive Moats

– Regulatory Armor: Visa sponsorships and immigration constraints keep a subset of workers captive. H-1B holders cannot easily leave without jeopardizing their status.
– Trust Shield: The ‘continuous tuning’ framing. As long as layoffs are presented as strategic adaptation rather than cost-cutting, social resistance remains muted.
– Physical Chains: Data center access, security clearances, and specialized infrastructure knowledge create narrow moats for specific roles. But these are being bridged by remote AI tools and vendor consolidation.

Future-Proofing Scorecard

| Timeline | Score | Commentary |
|———-|——-|————|
| 1 year | 2/10 | ‘Continuous tuning’ becomes standard practice. Workers acclimate to recurring layoff cycles as ‘the new normal.’ |
| 2 years | 1/10 | Talent flight accelerates as top performers realize loyalty is not reciprocated. Institutional knowledge hemorrhage. |
| 5 years | 0/10 | Big Tech operational workforces bifurcate: small elite teams and gig contractors. The middle is gone. |
| 10 years | 0/10 | The concept of ‘tech worker’ as a stable career category has dissolved. Employment is project-based, vendor-mediated, and AI-supervised. |

The Verdict

The article documents the normalization of layoffs without quite grasping what it is documenting. The framing is still sympathetic to management: ‘continuous tuning’ as necessary adaptation, uncertainty as a justification for cuts, competitive pressure as an irresistible force. The Oracle is less generous.

The verdict is that ‘continuous tuning’ is not a management strategy—it is an admission of strategic failure. These companies do not know what AI will do to their businesses, so they are cutting workers preemptively to buy narrative cover. The cuts are not enabled by AI capability. They are enabled by AI storytelling. Boards demand ‘AI-driven productivity gains,’ so management delivers them by removing the people who could have produced actual gains.

Moyan Chen, the Meta data scientist who was laid off, described the experience as ‘more like relief than pain.’ This is the most damning detail in the article. When workers are relieved to be fired, the employment relationship has ceased to function. The ’28 days of hell’ between announcement and execution—the period Meta workers endured—is not an aberration. It is the prototype. Continuous tuning means continuous anxiety, continuous precarity, and eventually, continuous exit.

The tech industry built its mythology on ‘talent’ and ‘innovation’ while quietly optimizing for headcount fungibility. AI delivers the excuse to complete the optimization. The layoffs are not the end. They are the tuning fork striking the note that the rest of the industry will follow. The verdict: the employment model is not broken—it is being replaced by something that does not need employees.