2026 Layoffs and AI: How to Reduce Operating Costs Without Losing Service Quality

How many 2026 job cuts cited AI?

Employers cited artificial intelligence in announcements covering 121,401 of the 617,189 US job cuts announced from January through October 2026, about 20% of the total, according to the Challenger, Gray & Christmas October 2026 job cuts report (opens in a new tab), published November 5, 2026. Figures in this article reflect reports available as of November 6, 2026.

Overall cuts are well below last year, and AI's share is shrinking. AI was cited for 1,265 cuts in October, its lowest monthly total this year, and market and economic conditions have overtaken AI as the leading reason for cuts in 2026.

Read these numbers with two limits in mind. First, Challenger counts announced job cuts, which may be phased in over time, reduced or not carried out. Second, the reason is the one the employer cited: a company citing AI does not prove that AI alone caused those job losses. Challenger also tracks automation separately from AI, under "Technological Update," which accounts for far fewer cuts this year.

Which reasons and industries account for 2026 job cuts?

Market and economic conditions, artificial intelligence, closings and restructuring are the four largest reasons employers cited for job cuts through October 2026. Demand downturns are smaller but rising fast: 21,347 cuts so far, up 145% from 8,701 in the same period of 2025.

By industry, technology companies announced the most cuts by a wide margin: 171,501 so far this year, about 28% of the total and up 21% from the same period of 2025.

For a business owner, the reason behind a cut changes what comes next. A cut driven by falling demand or a closure usually means less work. A cut tied to AI or restructuring often means much of the same work still has to be done a different way, and that is where service quality is won or lost.

What happens to the work after a layoff?

Some of it goes away and some of it stays. When cuts follow a closure or falling demand, there is less work to do. When they follow restructuring or new technology, customer inquiries still arrive, accounts still need reconciling, IT issues still need fixing and patients still need help booking appointments. Without a new operating model, that remaining work lands on whoever is left.

When positions are removed without redesigning the workflow, businesses commonly see:

• Longer response and resolution times for customers
• Heavier workloads and more overtime for the remaining team
• Delays in billing, reporting and administrative processing
• Less consistent service and lower customer satisfaction
• More operational risk, because knowledge sits with fewer people and processes become fragmented

How the change is handled matters too. Clear communication with the people leaving and the people staying, realistic workloads and documented handovers protect service as much as any tool does. A smaller team is not automatically a more efficient team; the savings only hold if the work has been reassigned on purpose.

What can automation and AI take over, and what still needs people?

Routine automation and AI-assisted processing both take repetitive work off a team, but they work differently. Routine automation follows fixed rules: if an invoice matches its purchase order, post it; if a form is complete, route it. AI-assisted processing handles less structured input, such as classifying an email, drafting a reply, summarizing a conversation or pulling fields from a scanned document. Its outputs can be wrong in ways rules-based automation is not, so they need sampling and review.

Work that needs judgment should stay with people: a complaint from an upset customer, an unusual financial discrepancy, or a patient concern that needs escalating. NIST's AI Risk Management Framework (opens in a new tab) was developed to help organizations manage the risks AI poses to individuals, organizations and society, and human oversight is a core part of managing those risks.

The goal is to automate the tasks that suit automation and keep clear ownership of every decision and service outcome. Our guide on deciding whether a task needs people, automation or AI walks through that call task by task, and what people should own in an AI workflow covers the oversight side.

Where does outsourcing fit in a leaner operating model?

For some businesses, outsourcing adds capacity and specialist skills without rebuilding every role in-house. A company can keep a small internal management team and use an external team for customer inquiries, administrative processing, technical troubleshooting or bookkeeping. Whether that makes sense depends on workload, complexity, security and service requirements.

The outsourcing industry is adjusting to AI as well. In its 2028 outlook released July 14, 2026 (opens in a new tab), the IT and Business Process Association of the Philippines (IBPAP) forecast industry revenue of $43.3 billion to $50.5 billion and 1.85 million to 2.14 million full-time employees by 2028, describing a shift "from capacity to capability." These are forecasts. The Philippine Star reported (opens in a new tab) that the range is below the $59 billion revenue target in IBPAP's 2022 roadmap, and that IBPAP president and CEO Jack Madrid said AI had affected some entry-level jobs in the industry and those employees had been redeployed to new work.

A practical operating model puts each type of work with the resource best suited to it:

This is not about replacing every employee. It is about matching the work to the right resource and keeping someone accountable for the result. See how we approach outsourcing and business support on that basis.

Which business functions should you review first?

Start with functions that have measurable volume, recurring workflows and clear performance indicators, because those are the easiest to price, pilot and compare.

Not every function should be outsourced or automated. Weigh the savings against quality, compliance, information security and the effect on customers, and check which privacy rules apply before any patient, employee or financial data leaves your systems.

How do you restructure operations without hurting service?

Map the work and its true cost before you change staffing or buy another AI tool, then test the new model on a small scale. These five steps keep the decision grounded in data.

Measure the pilot against your baseline, using the same definitions before and after the change:

• Cost per transaction: total cost of the work, including people, software, supervision and rework, divided by completed tasks.
• Quality and errors: error rate, rework rate and how many issues customers or auditors catch.
• Turnaround: time to first response and time to resolution or completion.
• Customer satisfaction: satisfaction scores, complaints and repeat contacts on the same issue.
• Human oversight: the share of automated or AI outputs a person reviews, and the escalation rate for exceptions.

Before handing any task to an outside team, document it properly. Our checklist of what to prepare before outsourcing a recurring task covers the handover.

What are the most common mistakes when cutting costs with AI?

The most common mistake is cutting first and redesigning later. By the time the gaps show up in customer complaints, the people who knew the process may already be gone.

• Measuring success by payroll savings alone, while rework, delays and churn rise elsewhere.
• Automating a broken process, which makes the same mistakes faster.
• Launching AI without a named owner who reviews outputs and handles exceptions.
• Outsourcing a task without documentation, quality targets or a way to escalate.
• Skipping the pilot and switching everything over at once.

If a task is rare, highly judgment-based or tied to a sensitive relationship, it may be cheaper overall to keep it exactly where it is.

How does FilWeb Asia help businesses through operational change?

FilWeb Asia Inc., based in San Pedro City, Laguna, Philippines, supports US and international businesses with outsourced customer support, technical support and help desk, bookkeeping, HR administration, healthcare administrative support and back-office and data operations, alongside AI workflow assessment and automation. We start by understanding your current operations and goals, then combine trained professionals, documented processes, oversight and the right amount of automation.

This article provides general business guidance, not employment or legal advice.

Frequently asked questions