Microsoft Offers 60,000 Buyouts as AI Layoffs Reshape Workforces

Microsoft is reportedly offering buyouts to 60,000 employees as it trims costs amid growing AI adoption, following an earlier voluntary retirement scheme that covered 7% of its US workforce. That figure, reported this week, is the sharpest single data point in what has become one of the most contested questions in workforce management: whether cutting headcount in anticipation of AI capability actually pays off.

The buyouts are expected to touch some of Microsoft's best-known divisions — Xbox, Windows, and Copilot — while cybersecurity, datacenter, and marketing teams are reportedly excluded.

What separates a smart AI layoff from a costly one?

Lukas Alexander, VP of Customer Success at ChurnZero, told CX Today this week that leaders need to know whether an AI-led workforce protects, improves, or damages customer outcomes before acting.

"Real genuine AI-driven efficiency means that teams can manage the same or more revenue with fewer resources and can maintain, or even enhance, the customer experience with the same or fewer resources," he said.

"The one thing that stayed the same … is that customers still need to see outcomes. I think that needs to be the measure."

The distinction matters because the track record so far is mixed. In numerous cases, vendors overestimated the results of mass layoffs and rehired only months later. Forrester's 2027 predictions expect CX staffing to rise by 20% by the end of next year due to overcutting — meaning some of today's cuts may reverse within 18 months.

Part of the productivity gap comes from weak deployment: only 45% of CX decision-makers' teams actively use AI tools. Cutting capacity before measuring the effect on customer outcomes, analysts warn, risks shifting work rather than eliminating it.

The takeaway: companies that don't separate automatable tasks from work needing human oversight end up buying bottlenecks and burnout, not efficiency.

How is Barclays balancing cuts and commitments?

Barclays, the first major UK bank to make the AI-layoff plunge, expects to remove 5,000 full-time roles by the end of this year — 5% of its workforce. That follows 1,000 role reductions earlier in the year.

Roughly 60% of the cuts will fall on support functions, including developer roles, reflecting both organizational simplification and early AI productivity effects. The bank will reinvest on the retention side: it plans to prioritize internal candidates for vacancies and invest £150 million in skills by 2030.

For individuals in affected functions, that reinvestment commitment is the practical difference between an exit and a possible internal move. But leaders copying Barclays need a clear view of which tasks get automated, which roles get redesigned, and which skills the future org chart actually requires.

Where is AI adding capacity instead of removing people?

Zendesk took the opposite path. Its new Employee Service AI Agents and Advanced Action Flows take employee requests "from question to outcome" — enhancing existing staff rather than replacing them.

Shashi Upadhyay, President of Product, Engineering, and AI at Zendesk, said: "AI should do more than answer employees' questions – it should help get their work done the way they intended to."

"Too often, employees lose time chasing approvals and navigating internal systems instead of doing the work they were hired to do. That's the promise of an agentic employee experience: less friction for employees and more time for the work that matters."

The numbers are concrete: customers drove 6 million workflow executions in the first five weeks after accessing Advanced Action Flows, saving roughly 500,000 hours of manual work — about five minutes per execution. That's time recovered rather than headcount removed, and organizations can validate it against their own error rates and employee outcomes.

Also this week

• RingCentral extended workforce forecasting and quality management to AI agents, adding live monitoring, quality scoring, journey analytics, and staffing forecasts for AI workers, plus AI-to-human handoff tracking under shared quality standards.
• Fountain's analysis of 12.5 million US frontline job applications found only about 8% resulted in hires, with 881,000 applications timing out without a clear decision in 2025. AI's share of first interviews jumped from 2% to 14% in one year.
• Genesys earned Great Place to Work certification across 23 countries, up from 17 in 2025, with 92% of surveyed employees calling it a great workplace and 90% reporting professional development opportunities.

For frontline job seekers, that Fountain figure means roughly one hire per twelve applications — and a first interview that's now seven times more likely to be conducted by AI than a year ago.

The pattern across the week is consistent: where AI deployments are measured against outcomes, capacity grows; where they're assumed, staffing boomerangs — and with Forrester projecting that 20% CX staffing rebound, the cost of cutting first and measuring later is about to become visible.