Microsoft fires the first shot in layoffs! AI intelligence triggers a wave of unemployment in 2025, Silicon Valley giants halt hiring programmers

In December last year, OpenAI announced significant advancements in coding with o3.

Some CS graduates feel like their major was a waste:

CS graduates who found relevant jobs might still be relieved, after all, o3 is only open to security researchers and has not yet entered the market, and its cost-performance ratio might also be relatively low.

But this time, the "wolf" has really come!

The "Cold Winter" of the American Job Market

Before OpenAI announced the new AI model o3, cloud computing giant Salesforce announced that it would no longer hire software engineers by 2025.

In 2024, Salesforce employees have already felt the "chill"—after all, this company is known for its layoffs. In January of that year, Salesforce laid off 700 people—about 1% of its 70,000 global employees.

For many affected individuals, this is heartbreaking, but compared to the previous year, it is just a drop in the bucket.

In January 2023, this cloud computing giant laid off 7,000 people—about 10% of the entire company. At that time, CEO Marc Benioff wrote a letter to employees stating that this was due to "hiring too many employees during the COVID-19 pandemic."

Salesforce's layoffs are just a microcosm of the layoffs in the U.S. last year.

According to data from Layoffs.fyi, after mass layoffs in 2022 and 2023, over 150,000 jobs were cut globally in 2024, involving 545 companies.

Large companies such as Tesla, Amazon, Google, Snap, and Microsoft conducted mass layoffs in 2024.

Some smaller startups also experienced layoffs, with some even shutting down completely.

In 2025, layoffs will continue, with large American companies including Microsoft, BlackRock, and Ally already confirmed to lay off employees.

Major Layoffs in the U.S. Tech Industry in 2025, Microsoft Leads the Way

Since 2022, the companies with the most layoffs have been Intel and Tesla. These two companies laid off 15,000 and 14,000 people respectively last year, becoming the employers with the highest layoff numbers in recent years.

Despite the varying reasons for company layoffs, these cost-cutting measures are closely related to the backdrop of technological transformation.

According to a recent report by the World Economic Forum (WEF), it is expected that about 41% of companies globally will reduce their workforce due to the rise of AI in the next five years.

Previously, companies such as Dropbox, Google, and IBM have announced layoffs related to AI.

The past year has passed, and as the new year begins, a larger wave of tech layoffs is imminent.

Microsoft

First and foremost is Microsoft.

BI exclusively reported that a Microsoft spokesperson has confirmed that the company will soon initiate its 2025 layoff plan, although the specific number of layoffs has not been disclosed.

However, in this layoff plan, Microsoft will focus on employees with poor performance.

It is reported that layoffs will occur across the entire company, especially in the security department.

The spokesperson stated, "Microsoft focuses on high-performance talent. We have always been helping people learn and grow, and if they continue to perform poorly, we will take appropriate measures."

The employees who are let go due to poor performance will be replaced by new hires, according to Microsoft.

Therefore, the total number of employees in the company will not change significantly. As of the end of June last year, Microsoft had a total of 228,000 full-time employees.

Insiders have indicated that, like its competitors, Microsoft is currently taking a tougher stance on performance management.

In recent months, the company's management has been evaluating employees, including those at level 80, which is one of the highest positions in the company.

In fact, as early as the end of last year, various rumors were circulating on the anonymous social platform Blind.

Some users claimed that Microsoft would conduct large-scale layoffs on January 15, 2025, with a potential rate as high as 5%.

Aqua Security

Aqua Security is a company focused on cloud-native security protection platform solutions.

As part of a strategic restructuring, the company plans to lay off dozens of employees in the global market to improve profitability.

Ally

A spokesperson for digital financial company Ally confirmed that the company will cut about 500 employees (out of a total of 11,000 employees) and has begun sending emails to employees on Tuesday A spokesperson stated, "As we continue to adjust the company's scale, we have made a difficult decision: to selectively reduce staff in certain areas while continuing to hire in other areas of the business."

Ally will provide severance pay, job placement support, and opportunities for employees to apply for internal vacancies.

Altruist

Los Angeles-based fintech company Altruist plans to lay off 37 employees, affecting about 10% of its workforce.

Additionally, according to statistics from the Trueup website, a total of 16 tech companies have laid off 2,977 employees so far. Among them, it lists Microsoft with a layoff count of 2,280.

2,977 employees is just the beginning of layoffs in 2025.

Will AI Agents Trigger a "Wave of Unemployment"?

In the next 2 to 4 years, netizens predict that AI agents may lead to a depression-level wave of unemployment.

How exactly will this happen?

One-fifth of white-collar workers have very simple jobs, and "powerful AI" could quickly replace them:

Step 1: The company installs monitoring tools such as keyloggers on employees' computers.

Step 2: AI observes and learns employees' work behaviors.

Step 3: AI replaces humans (note that AI can already operate computers, clicking and typing with ease).

You might ask, "But what about those special cases?"

For certain jobs, AI may not be able to complete 100% of the tasks, but it can accomplish 80%, which means the company can lay off 80% of the employees.

Typically, the widespread adoption of technology takes decades, as reshaping supply chains requires a long time.

But that is not the case here—AI agents only need to learn by observing employees' screens.

Perhaps we can find new jobs for the displaced employees, but maybe we cannot?

Because unlike previous economic revolutions, the other jobs they could qualify for may also be simple enough to be replaced by AI agents.

Please note, AI will continue to become exponentially smarter.

Historically, horse-drawn carriages have been replaced by cars, and this time, human jobs may be replaced by AI.

Salesforce is hiring salespeople to promote AI products

Salesforce's CEO and founder Marc Benioff revealed that due to the significant productivity improvements brought by AI technology, Salesforce has decided to freeze hiring for software engineers in 2025.

Moreover, due to AI agents, Salesforce will also reduce the number of after-sales engineer positions.

When asked whether Salesforce will have more or fewer employees in five years, he stated that the company "might have more."

However, he further explained: "We will not hire more software engineers next year because this year we have increased productivity by over 30% through Agentforce and other AI technologies used for engineering teams—engineering development speed is incredibly fast."

"In addition, next year we will reduce the number of support engineers because we have agents. We will increase the number of salespeople next year because we need to explain in detail the value we can achieve through AI. Therefore, in the short term, we may add 1,000 to 2,000 salespeople."

Salesforce has already applied AI agents in the customer support field.

Experienced workers can provide clear and detailed instructions to effectively utilize the agents.

However, newcomers may make inappropriate choices, and AI agents may instead become technical debt.

Can AI agents replace programmers?

Salesforce's approach seems to suggest that agents cannot yet replace salespeople and customer service representatives, but they can replace software engineers.

According to the latest survey by the World Economic Forum, 86% of respondents believe that the macro trends driving business transformation include AI and information processing technology.

But the truth is not simple.

The CEO of Lazy AI does not believe that current AI can replace software engineers and provided six reasons.

1) Agent-based programming = False advertising

50% of AI-generated code is worthless.

The reason is: every time the code is modified, there is a 50% chance of introducing some kind of bug or unexpected behavior, which may be triggered by the prompts.

If a large language model (LLM) makes 5 changes at once, it can introduce multiple bugs upon completion, and it may not even know how to trigger these bugs.

2) General coding tools are a lie

When a company claims to have built an AI product that is framework-agnostic and application-type agnostic, requiring only prompts to let AI do all the work, you should remain skeptical.

Try it out, and you will find that AI sometimes writes Python in JS files, confuses dependencies, and mixes comment syntax.

Most companies cannot solve these types of problems.

3) Authentication is your responsibility

In AI coding tools, run the following prompt: "Create a Google login page that redirects to my profile page."

Broken OAuth flows, redirect URI errors, links pointing to documentation, and wasting time in the Google Cloud Console for API keys...

Thousands of users have left in frustration because of this.

4) LLMs cannot see your database

Almost every project requires a database. However, the database is constantly changing during project development.

The following cycle is common:

1. Change the database, and the AI doesn't know what to do.
2. Change the code, and the database errors out.
3. Need code migration, and the AI writes a migration script with bugs.
4. Then it keeps repeating.

5) Pay now, never succeed

You may have experienced the following situation:

When you ask the AI to build a new feature, it deletes parts of the project. The AI copies certain parts of the code. When you explicitly ask it to do X, it does Y instead. The AI gets stuck fixing the bugs it introduced itself.

When you are paying for usage, it is not enough for the AI model to be "getting better."

A misplaced parenthesis can ruin an entire page. Every time a new AI model is released, it requires weeks of testing with real-world applications.

6) Engineering itself is hard.

As for wanting to use AI to build complex projects, it is a pipe dream.

Source: [New Intelligence](https://mp.weixin.qq.com/s?__biz=MzI3MTA0MTk1MA==&mid=2652556919&idx=1&sn=1d563f6e1bc7b6731b67a31d5a9ed3cc&chksm=f0903bf1c245cb63920e5f1838726d6847fb3b4be11b6e906e6b4d2bdbb4828ab46adfa2b581&mpshare=1&scene=23&srcid=0109ojFhEcAesK6ag1uMp5iz&sharer_shareinfo=40ed420782b4688fe8c75e5e5fbe68a4&sharer_shareinfo_first=40ed420782b4688fe8 c75e5e5fbe68a4#rd), Original title: "Microsoft Fires the First Shot in Layoffs! AI Intelligence Triggers a 2025 Unemployment Wave, Silicon Valley Giants Halt Hiring Programmers"

Risk Warning and Disclaimer

The market has risks, and investment requires caution. This article does not constitute personal investment advice and does not take into account the specific investment goals, financial situation, or needs of individual users. Users should consider whether any opinions, views, or conclusions in this article align with their specific circumstances. Investment based on this is at one's own risk