Today's brief · 27 October 2026

Microsoft offers buyouts to about 60,000 staff and OpenAI halves the price of its computer-use model as AI squeezes payrolls and cuts the cost of automation.

Developments reported between 10:30 PM IST on 26 October and 2:30 AM IST on 27 October 2026. Two verified stories on how Microsoft is using a large voluntary exit program to contain costs while it funds AI spending, and how OpenAI is cutting the price of AI that can operate software. Each is written in plain language with the key numbers, the sectors most affected, and specific recommendations for the people who will need to act on it.

Stories in this brief
2 developments
Sectors covered
2 verticals
Read time
7 minutes
Filter by sector
2 of 2 stories visible
01 / 02
Operations & Enterprise

Microsoft offers voluntary buyouts to about 60,000 employees, more than a quarter of its global workforce, as it works to contain costs and fund AI spending

Microsoft said on Monday that it is opening a time-limited, voluntary exit program to roughly 60,000 staff, according to Reuters, which notes that Business Insider first reported the plan. Reuters says the aim is to keep costs in check and help pay for the company's AI investments. According to Microsoft, the offer covers employees at level 70 and below in the United States, Canada, India and the United Kingdom. A person familiar with the matter told Reuters that the eligible groups include AI experiences, Copilot, finance, human resources, Xbox, and Windows + Devices, while marketing, the commercial business and security are reportedly left out. In a memo to staff, chief people officer Amy Coleman described the program as "a new choice to make that transition now with support."

The offer is much wider than the company's earlier moves this year. Reuters reports that Microsoft offered buyouts earlier in 2026 to about 7% of its US workforce, or roughly 9,000 people, and that in July it announced 4,800 job cuts, about 2.1% of its global staff. The company had 223,000 full-time employees worldwide as of 30 June 2026, according to its annual filing, and Reuters says it is spending heavily on AI data centers and chips, with capital expenditure forecast at $175 billion for calendar 2026. Microsoft says the new program builds on its earlier voluntary retirement plan. It reports first-quarter results after the market closes on Wednesday.

Why it mattersThis is one of the clearest cases yet of a large technology company reshaping its workforce to pay for AI infrastructure. Buyouts let people leave by choice rather than through layoffs, but at this scale they can still thin out teams and slow work in the units involved, which reportedly include Copilot, Windows and Xbox. For customers and partners, the signal is that Microsoft is putting capital into data centers and chips first, and that the people they deal with in some product and support teams may change over the coming months.
Key numbers
About 60,000
Employees offered voluntary buyouts, per Reuters
Over a quarter
Share of Microsoft's global workforce covered by the offer
223,000
Full-time employees worldwide as of 30 June 2026
$175 billion
Forecast capital spending for calendar 2026
About 9,000
US employees offered buyouts earlier in 2026, about 7% of the US workforce
4,800
Job cuts announced in July, about 2.1% of global staff
Sector impact and what to actually do
For CIOs and vendor managers with large Microsoft estates

Over the next 30 days, list your named Microsoft account, partner and support contacts in Copilot, Windows and related groups, and ask your account team in writing how continuity will be handled if those people take the offer.

For CHROs and workforce planners at other large companies

Before your 2027 budget is set, model what a voluntary exit program tied to AI spending would cost and who would take it, so you can weigh it against layoffs and hiring freezes if your board asks how you will fund AI investment.

For equity analysts and investor relations teams covering software

Ahead of Microsoft's results on Wednesday, prepare questions on how many staff accept, the expected one-off cost and savings, and how the program relates to the $175 billion capital spending forecast.

Source Reuters (via Yahoo Finance) · 26 October 2026
02 / 02
Operations & Enterprise Cybersecurity & Infrastructure

OpenAI releases GPT-6.1 Terra, a faster model for computer work priced at $1 per million input tokens, half the API price of GPT-6.1 Sol

OpenAI on Monday released GPT-6.1 Terra, which it presents as a quick, efficient model for visual reasoning, operating software on a computer and clearer writing. It succeeds GPT-5.6 Terra as the company's lower-cost option. According to OpenAI, the model is available now to Plus, Pro, Enterprise and Edu users in ChatGPT Work and in Codex, and to developers through the API as gpt-6.1-terra. Standard API pricing is $1 per million input tokens, $0.05 per million cached input tokens and $5 per million output tokens, which OpenAI says is 50% lower than both GPT-5.6 Terra and its larger GPT-6.1 Sol model.

On OpenAI's own benchmarks, Terra comes close to the company's bigger models on many tasks for much less money. OpenAI says it completes tasks more than three times faster than GPT-6 Astra and GPT-6.1 Sol, and that at maximum reasoning effort it solves 50.3% of tasks on the Terminal-Bench 4.0 test. On an internal test built to provoke mistakes, the company reports that the error rate at low reasoning effort dropped to 16.7% from 53.6% for GPT-5.6 Terra, and reached 4.3% at maximum effort. OpenAI also says Terra meets the "Critical" threshold for cybersecurity under its Preparedness Framework and carries over the safeguards developed for GPT-6 Astra. For the most demanding coding work, the company still recommends Sol and Astra. All of these results are OpenAI's own figures.

Why it mattersFor companies running AI agents that click through software, fill in forms or handle IT tasks, cost and speed per completed task matter more than top benchmark scores. A model that OpenAI says is half the API price of Sol and finishes tasks several times faster changes the business case for automating routine back-office work. The "Critical" cybersecurity rating is the other side of the same capability: a model good at operating systems can also be misused against them, so access and oversight need as much attention as savings.
Key numbers
$1
Standard API price per million input tokens
$5
Standard API price per million output tokens
50% lower
Standard API prices versus GPT-5.6 Terra and GPT-6.1 Sol, per OpenAI
Over 3x
Faster task completion than GPT-6 Astra and GPT-6.1 Sol, per OpenAI
50.3%
Terminal-Bench 4.0 tasks solved at maximum reasoning effort
16.7%
Error rate at low effort on OpenAI's internal error-inducing test, down from 53.6%
Sector impact and what to actually do
For CIOs and heads of automation running AI agent pilots

Within the next 30 days, rerun one existing agent workflow, such as invoice entry or ticket triage, on Terra and compare cost per completed task and error rate with your current model before renewing any model commitments.

For CFOs and FinOps leads managing AI spend

Before 2027 budgets are locked, reprice your forecast AI usage at the new $1 and $5 per million token rates, and ask software vendors that build on OpenAI whether they will pass the lower cost through in their pricing.

For CISOs and IT security leads

Within two weeks, review which staff and tools can use computer-use models in ChatGPT Work and Codex, and confirm that agents able to operate company systems run with limited permissions and full logging, given OpenAI's own Critical cybersecurity rating.

Source OpenAI · 26 October 2026

No stories tagged for this sector today.

Try another sector or return to the full view.

Editor's note
The AI build-out is now visible on both sides of the ledger: companies are trimming payroll to pay for AI, while the price of AI that can do routine computer work keeps falling.

Microsoft's offer of voluntary buyouts to about 60,000 employees, more than a quarter of its global workforce, comes as it forecasts $175 billion of capital spending for calendar 2026 and spends heavily on AI data centers and chips. Earlier this year it offered buyouts to about 9,000 US staff and announced 4,800 job cuts, so the new program is a step up in scale rather than a change of direction. On the same day, OpenAI released GPT-6.1 Terra, a model for operating software that it prices at $1 per million input tokens and $5 per million output tokens, half the API price of its larger Sol model, and that it says finishes tasks more than three times faster. Read together, the two stories show the trade-off many boards now face: money for AI infrastructure has to come from somewhere, and payroll is one of the few lines large enough to matter, while the tools that automate routine office work are getting cheaper and quicker. For executives, the practical step is to test what cheaper agent models can actually do in their own workflows, and to prepare a clear answer to how AI investment, and its effect on staffing, will be funded in 2027.

27 October 2026