Google launches a global voluntary exit program with up to 12 months’ pay; Hasabis: the future…
Google Launches Global Voluntary Resignation Program, Up to 12 Months’ Salary Compensation! Hasabis: The Future…
Summary
Author | Dongmei – A tech giant with quarterly revenue exceeding $120 billion and cloud revenue doubling year‑over‑year, why is it still paying employees to leave? This seemingly contradictory scenario is playing out at Google. On November 4, according to …
A tech giant with quarterly revenue exceeding $120 billion and cloud revenue doubling year over year—why is it still spending money to let employees go?
This seemingly contradictory situation is playing out at Google.
On November 4, Reuters reported exclusively that Google, under Alphabet, has expanded its voluntary departure program this year to multiple countries including Canada, France, Germany, Japan, and Switzerland. Eligible employees can voluntarily apply to leave and receive corresponding financial compensation. In Switzerland, the compensation is one month’s salary for each year of service, up to a maximum of 12 months.
What’s more noteworthy is that this adjustment does not occur during a period of poor performance at Google.
Just a week ago, Alphabet just delivered a strong quarterly report: Q3 2026 revenue reached $127.6 billion, up 25% year‑over‑year; Google Cloud revenue surged 103% year‑over‑year to $30.7 billion; operating profit reached $40.6 billion, up 30% year‑over‑year.
But it must be said, another set of numbers is equally striking: Alphabet expects to spend $195 billion to $205 billion this year on capital expenditures, with $51.4 billion spent in Q3 alone. Meanwhile, the company’s free cash flow turned negative that quarter, with a net outflow of about $9.1 billion.
On one hand, revenue continues to grow; on the other, AI investment keeps expanding. Google is undergoing a new resource reallocation.
AI and cloud computing divisions remain hiring priorities, while some traditional business teams have begun offering severance to employees.
According to Reuters, a Google spokesperson said that in 2025 some teams began implementing a voluntary departure program with severance pay, and in June of this year the program was expanded to several other countries to support the company’s next phase of important work.
This is not the first time Google has implemented a similar program.
Since 2025, Google has rolled out voluntary departure arrangements in certain business units in the United States and the United Kingdom. Management previously explained that the program mainly targets employees who feel they no longer fit the company’s pace of development or strategic direction.
Unlike direct layoffs, voluntary departure programs typically allow eligible employees to apply proactively, but approval still depends on the company’s specific policies.
This report reveals that the compensation arrangement in Switzerland is particularly noteworthy.
Switzerland is a key engineering and R&D hub for Google in Europe, with about 5,000 local employees. Three sources told Reuters that the voluntary departure program covered multiple business units in Switzerland but did not include Google DeepMind.
Google announced the plan in mid‑June and provided feedback to applicants by mid‑July. Some employees who accepted the offer began leaving the company starting in August.
According to the disclosed local compensation standards, employees receive a payout equivalent to one month’s salary for each year of service, up to a maximum of 12 months. One source said that in a team of roughly 50 people, at least four have taken the departure offer.
However, it is still unclear how many employees are participating in the latest round of voluntary departure plans.
It is worth noting that Google previously carried out staff adjustments in Switzerland. In 2023, the company cut more than 200 employees there, sparking protests and strikes.
Reuters also pointed out that at least one Google department conducted further layoffs after implementing a voluntary departure plan.
For Google, the current issue is clearly not a lack of revenue, but the scale of AI-era investment is reshaping the company’s financial structure.
On October 28, Alphabet released its third‑quarter financial results for the period ending September 30, 2026. In terms of revenue and operating profit, Alphabet performed quite strongly. InfoQ compiled Alphabet’s core financial data for Q3 2026 based on the report:
It should be noted that the decline in net profit is also affected by non‑operating investment gains and losses, and cannot be attributed solely to AI spending.
Google CEO Sundar Pichai said during the earnings call that the company’s revenue has reached an annualized run rate of about $500 billion, delivering double‑digit growth for the 13th consecutive quarter.
Among them, Google Cloud has become the most prominent growth engine.
In the third quarter, Google Cloud revenue reached $30.7 billion, up 103% from $15.2 billion a year earlier; operating profit hit $11.1 billion, more than double the $3.6 billion from the same period last year.
This growth is mainly driven by enterprise AI infrastructure, enterprise AI solutions, and traditional cloud services, while TPU system sales also contributed a significant increment.
However, a doubling of cloud revenue does not mean Google has solved all the challenges of AI investment.
The earnings report shows Alphabet’s Q3 capital expenditures reached $51.4 billion, mainly for servers, networking equipment, and data center infrastructure. In the same quarter, the company’s operating cash flow was $42.3 billion, while free cash flow was about negative $9.1 billion.
In other words, although Google’s day‑to‑day operations still generate a lot of cash, in Q3 its capital spending already exceeded the cash generated by operating activities.
Google CFO Anat Ashkenazi further confirmed that full‑year 2026 capital expenditures are expected to stay between $195 billion and $205 billion, with a significant increase projected for 2027. This suggests that the cash‑flow pressure from AI infrastructure expansion will not disappear quickly.
But Google did not launch a voluntary departure program because its cash is about to run out. Its balance sheet still holds substantial cash and financial assets, and its business profitability remains strong. However, as capital spending continues to expand, management needs to be more cautious in controlling other costs.
On October 29, Reuters, while analyzing the AI investment frenzy among tech giants, noted that Wall Street is re‑examining how these companies allocate capital. In the past, investors were willing to reward firms that actively built AI compute power and expanded infrastructure investment; now the market is asking when those investments will generate sufficient cash returns.
Alphabet felt this pressure. On October 29, after the earnings release, the company’s stock plunged, at one point falling more than 12%, wiping out about $500 billion in market value. The market’s concern is not only the scale of AI investment but also whether the search advertising business that underpins this investment can continue to grow.
So Google is expanding its voluntary resignation program at this time, possibly not because of deteriorating company performance, but more like re‑examining other costs while rapidly expanding its AI business.
Interestingly, even though Google is investing in AI on an almost unprecedented scale, its core AI business head Demis Hassabis has a nuanced attitude toward this competition.
In a recent official podcast, Hassabis admitted that if he could steer AI research as he wished, he would have preferred the technology to stay in the lab longer.
According to his original vision, DeepMind could cautiously advance AGI research while applying the technology to science and medicine, creating more results like AlphaFold.
But the commercial success of generative AI changed all that.
Hassabis admits that products like ChatGPT have let the public truly experience AI’s value and have attracted massive funding and research resources to the field. Yet what follows is increasingly fierce competition among tech companies and even nations.
He describes it as possibly one of the most intense capitalist competitions in history. Some investors who lived through the internet bubble told him that the current AI industry competition intensity could be ten times that of back then.
“This makes it difficult for us to conduct rigorous scientific research at the same time,” Hassabis said.
For Google, this competition is turning into tangible capital expenditures.
In 2026, Alphabet expects to spend $195 billion to $205 billion on capital expenditures, mainly supporting infrastructure such as servers, networking equipment, and data centers. At the same time, Google continues to vie for AI research and engineering talent.
The reason Google is willing to invest so heavily in AI is that it sees the possibility of AGI arriving soon.
In the podcast, Hassabis reiterated his AGI timeline: by 2030, there is roughly a 50% chance that humanity will achieve artificial general intelligence.
He even believes that the societal transformation ultimately brought by AI could be ten times larger than the Industrial Revolution and happen ten times faster, and in the future one person might be able to do the work of ten people.
In Hassabis’ view, if AGI can help humanity solve major challenges such as energy, materials science, and nuclear fusion, the world could enter an era of immense material abundance within five to ten years after AGI’s emergence. He also said that simply scaling up existing models and improving the technology would make the probability of achieving AGI about 50%.
Pressure and excitement coexist, which will give rise to a kind of anxiety.
In this episode of the podcast, Hassabis also directly addressed the question of whether being at the forefront of AI development brings huge psychological pressure.
He admits that he sleeps very little, partly because he has too much work and partly because he finds it hard to fall asleep.
What excites him is that he is part of a technological revolution that could change human history; what worries him is that society does not seem ready yet.
Hassabis even believes that people are still underestimating the changes AI may bring in the next decade. AI could not only reshape the economy and employment but also force humanity to rethink the meaning of work, life, and even our own existence.
This statement, when applied to today’s Google, is rather thought‑provoking.
The productivity dividend brought by AI is still hard to quantify accurately. Yet an increasingly realistic question is: as companies begin to recalculate the value of each role, who will share the gains from technological progress, and who will bear the costs of transformation?
Google’s voluntary departure program does not directly prove that AI is replacing human employees. However, it shows that while tech giants race to bet on next‑generation technologies, organizational structures, talent needs, and capital allocation are also being re‑examined.
For companies in the AI era, the real challenge may not only be building more powerful models, but also ensuring that technological advancement stays in sync with human development.
Reference links:
https://www.theverge.com/tech/1026273/googles-offering-buyouts-to-more-of-its-employees
https://www.youtube.com/watch?v=B9u1kWv953A
https://x.com/LayoffAI/status/2118005600177021173
Disclaimer: InfoQ original, does not represent the platform’s views, does not constitute investment advice, and unauthorized reproduction is prohibited.
Harness is already very accurate, why still need an Agent Native Database
On December 18–19, the AICon Global Artificial Intelligence Development and Application Conference will be held in Beijing. This year’s conference focuses on exploring the boundaries of AI applications, featuring 13 technical tracks that cover core areas such as post-training for large models, inference optimization, agent engineering, knowledge engineering, intelligent agent evaluation and safety, self-evolving systems, AI software delivery, native architecture, and organizational transformation. Leading technical experts will gather to share breakthrough AI technologies and real-world practices for enterprise-scale deployment.
Get MirrorSpeed
Get MirrorSpeed