Microsoft's 60,000 Buyout Offers: Is It AI, or Is It Overhiring?

Microsoft has reportedly offered voluntary buyouts to 60,000 employees. That is roughly a quarter of a company that has already trimmed its workforce more than once this year. Is this the first big crack in the AI-era job market, or the tail end of a hiring binge that never made sense?

In a new video, Eli the Computer Guy reacts to the Guardian-sourced reports and argues for the second reading. This post pulls out the facts he cites, the numbers behind them, and his reasoning, and it labels anything that is my own inference.

Source video: "Microsoft 60,000 Employee Buyout Offers for AI Investment - Tech Jobs are Failing" by Eli the Computer Guy (28:49). All credit for the original commentary goes to Eli. This post summarizes his key points, and the opinions below are his unless marked otherwise.

What Was Reported

Eli opens with the headline as he read it: Microsoft reportedly plans to offer a buyout to 60,000 employees, in an effort to reduce costs amid the rising use of AI. He stresses that this comes after Microsoft has already laid off tens of thousands of people over the past couple of years.

The details he walks through:

He reads the company statement attributed to Amy Coleman, Microsoft's chief people officer: the offer "gives eligible employees who may be considering a future outside the company a new choice to make the transition now with support." Eli hears "a bit of an edge" in that wording, as if to say: wouldn't you rather leave now, with support, than later without it?

The live quote above is for Microsoft (NASDAQ: MSFT). It is context only, not a recommendation, and the video does not discuss the stock price.

His Argument: Overhiring, Not Just AI

Eli's central claim is that he would "argue this is still blaming AI" for what is mostly a hangover from interest rates and overhiring. His case has a few parts.

1. Tech plateaued. He says that as a long-time technology professional he sees little difference between what people use today and what they used about 15 years ago. Smart TVs and social media, for example, were already normal then. If a company is not shipping genuinely new things, he argues, it has to give investors something else to look at, and headcount growth is one way to do that.

2. Startup math. He recalls an angel investor who, around 2017, described backing 60 startups and expecting roughly this:

60 startups funded
  2 do really well   -> where the returns come from
 10 are adequate     -> probably do not lose money
 48 are expected to fail

Eli's point is that a great deal of staff time and resources was expected to be thrown away by design.

3. Remote-work scaling. He grants that moving work online meant cloud companies had to scale up. But he says they hired far more people than the job required, and he is blunt about the gap between "some" and "20,000." His illustration of why infrastructure needs fewer people than a service business:

Pizza shop (people scale roughly linearly)
  1 cook  ->  50 pizzas/hr
  2 cooks -> ~110 pizzas/hr
  4 cooks -> ~225 pizzas/hr

Servers (admins scale very slowly)
  1 server       -> 1 admin
  100 servers    -> 2 admins
  1,000 servers  -> 3 admins
  10,000 servers -> 4 admins

These are his illustrative numbers, not figures from any company.

4. Cheap money. He notes that hiring was inexpensive when interest rates were around 1%, and says what we are seeing now is an unwinding.

The Wider Layoff Picture He Cites

Eli places Microsoft in a longer list of cuts:

• Block: about 4,000 of roughly 9,000 employees, around 45%, according to Eli (he says he is going from memory on the figures).
• Oracle: 21,000 people.
• Meta: he refers to massive layoff numbers there as well.
• Microsoft itself: earlier layoffs of about 10,000 and then several thousand more, plus the April retirement offer, plus this buyout.

His own arithmetic: more than 10,000 early retirements already this year, plus about 5,000 expected buyout takers, would mean roughly 15,000 people leaving in 2026 on the offered programs alone.

What puzzles him is the shape of the cuts. Closing an office or winding down a failed product line makes sense to him, because those units are self-contained. Cutting a large fraction of staff across the board does not, since institutional knowledge about why systems were built a certain way and who the vendors are walks out the door. His twist is that if nothing of value leaves with them, they were not that valuable to begin with. That is a harsh opinion, and it is his.

Take the Buyout or Wait?

Eli shares a story about a family friend who received a buyout offer. Former senior managers told the friend that the choice was between taking a known package now or waiting a few months, when large-scale firings might start, with no guarantee of severance. Eli's own summary of the risk: if you take the buyout and find a job quickly, you are ahead, but if nobody hires you for a while, you can be in real trouble.

He also says he would not feel comfortable as a Microsoft employee right now.

Why Rolling Cuts Hurt

On management practice, Eli says it is better to prune once, decisively, and then tell the remaining staff the event is over, than to run repeated rounds. In his telling, employees are "skittish," and a constant drip of layoffs, retirement offers and buyouts leaves people with mortgages, kids and bills unable to focus. He argues that technical work takes a lot of brainpower, so this is the opposite of what you want from your engineers.

He contrasts this with the earlier era, when he says companies coddled developers with perks like unlimited paid time off because it was the only way to retain them.

Market Signals He Mentions

Near the end, Eli says the AI bubble is not admitted to be bursting, but points to a few numbers:

• CoreWeave, which he calls a "neo cloud" (basically an AI data center), is down about 60% over the year.
• Intel is down about 50% for the year.
• TSMC is down about 20% over the month.

He tells viewers to verify these themselves. Treat them as his claims at recording time.

My Take: Start Building AI Skills Now

This section is my own view, not Eli's. Whatever you think of the reasons behind these cuts, I keep coming back to one thing: AI tools are improving quickly, and they are getting better at more kinds of work every few months. I would rather be someone who knows how to use them well than someone who finds out too late that colleagues already do. In a job market that is clearly shifting, falling behind on AI skills is a risk you can avoid, and starting is cheap.

I am not making a claim about how many jobs AI will or will not replace. Nobody knows that. My point is simpler: the people who can work effectively with these tools will be in a stronger position, whichever way the debate goes.

Here is where I would start:

• Use an AI assistant in your daily work. Pick one real task each day, such as drafting, summarizing or debugging, and run it through an assistant. Habit matters more than any single trick.
• Learn to write good prompts. Be specific about the goal, give context and examples, state the format you want, and iterate when the first answer misses.
• Automate repetitive tasks. Look for chores you do weekly, like reports, ticket triage or data cleanup, and try scripts, workflows or agents to handle them.
• Understand the AI tools in your field. Find out what developers, admins, analysts or marketers in your area are already using, and try two or three of them.
• Verify the output. AI makes confident mistakes. Check facts, test code, and keep a human in the loop for anything that matters. Being the person who catches errors is a skill employers will value.
• Keep learning on a schedule. The tools change fast, so set aside a small block of time every week to try something new.

You do not need to become an AI researcher. You do need to be comfortable with these tools before your next job search, not during it.

My Takeaways (Inference, Not From the Video)

If you work in tech, here is a practical checklist inspired by the video. These are my suggestions, not Eli's.

• ☐ Keep your resume and portfolio current, whatever your employer says
• ☐ Build an emergency fund that covers several months, since Eli's own caveat is that re-hiring can take a while
• ☐ If you get an offer, read the terms and compare the package with what you would likely get if cuts came later
• ☐ Track skills that are tied to a product or revenue, since Eli argues that broad cuts do not discriminate by office or product line
• ☐ Pick one AI skill from my list above and practice it this week
• ☐ Be cautious about taking any single headline, "AI did it" or "overhiring did it," as the whole story

The Bottom Line

Eli's message is that the 60,000 figure is a symptom of a long period of overhiring, cheap money and slowing innovation, with AI as part of the explanation but not all of it. My own addition: whatever drives the cuts, keep your AI skills growing so you are not caught behind. Whether you agree or not, the numbers he lists are worth knowing, and the scale of the offer is hard to ignore. If you want his full take, including the stories and the closing note about his upcoming free classes and meetup group, watch the original video above and add your own thoughts in his comments.