By Vicky Sidler | Published 13 March 2026 at 12:00 GMT+2

Do you remember what you were doing in February 2020?

You were probably sitting in your office, planning a vacation, and assuming that a weird virus overseas was absolutely nothing to worry about. According to tech investor Matt Shumer, we are currently living through the exact same flavor of blissful, catastrophic ignorance regarding artificial intelligence.

He recently published a frantic warning to his friends and family, dropping the polite cocktail-party version of his job to tell them that the technology has crossed a terrifying new threshold. While we are all busy complaining about our morning commutes, a handful of researchers have released software that is fundamentally replacing human cognitive work.

Before you assume your cozy desk job is perfectly safe, we need to look at what the machines are actually doing behind closed doors.

TL;DR:

• Tech insiders are warning that artificial intelligence is now capable of doing complex white-collar work completely independently.

• The newest models are actually being used to build and debug the next generation of artificial intelligence, triggering a rapid intelligence explosion.

• Experts predict that fifty percent of entry-level cognitive jobs could be completely eliminated within the next one to five years.

👉 In a world where machines can do your technical work for free, your undeniably human expertise is your only remaining asset. If your marketing sounds like a robot wrote it, you are already losing sales. Identify the exact spots where your message is failing by downloading the 5-minute marketing fix.

Table of Contents:

Why Are We Completely Ignoring The Warning Signs?

To understand our collective denial, we have to look at how terrible the software used to be.

If you tried using a chatbot in 2023 or 2024 and thought it was just a confident nonsense generator, you were completely right. Those early versions hallucinated constantly. Because of that experience, most professionals assume the technology has hit a wall and is no longer a threat.

But judging the current state of artificial intelligence based on a free tool from two years ago is exactly like evaluating modern smartphones by trying to send a text on a plastic flip phone.

So what happens when you actually pay for the upgraded version?

What Is The Machine Actually Doing Right Now?

It is officially doing the jobs of the people who built it.

In February 2026, tech companies released massive new models like GPT-5.3 Codex and Opus 4.6. According to Shumer, he no longer does the actual technical work of his job. He simply describes an app in plain English, and the machine writes tens of thousands of lines of code, clicks through the buttons, tests the features, and fixes its own errors before presenting the finished product.

It is no longer just executing basic instructions; it is making intelligent decisions that require actual judgment and taste.

But what happens when a machine with good judgment turns its attention to its own source code?

How Did The Code Start Writing Itself?

This is the part that should make you sit up and pay attention.

When OpenAI released GPT-5.3 Codex, their technical documentation quietly admitted that this was their first model that was instrumental in creating itself. The artificial intelligence was actually used to debug its own training and manage its own deployment.

The machine is now intelligent enough to meaningfully contribute to its own improvement. Researchers call this an intelligence explosion, where each generation builds a smarter version of itself, which builds the next version even faster.

So if the machine is currently replacing software engineers, how long does your specific job actually have left?

Is Your Cozy Office Job Actually Safe?

Probably a lot less safe than your mortgage requires.

Dario Amodei, the CEO of Anthropic, publicly predicted that artificial intelligence will eliminate fifty percent of entry-level white-collar jobs within one to five years.

An organization called METR tracks how long an AI can work independently on complex tasks, and recent data suggests its capability is doubling every four to seven months. It is coming for law, financial analysis, medical reading, and consulting. This is not like factory automation, where you can just retrain for an office job, because the machine is getting exponentially better at all cognitive tasks simultaneously.

When the software can do the technical parts of your job perfectly, how do you convince a client to actually pay you?

Why Is Being Undeniably Human Your Only Survival Strategy?

Because authenticity is about to become the most expensive luxury in the business world.

If a robot can write a flawless legal brief or a complex financial model in fourteen seconds, your clients are no longer paying for your technical output. They are going to pay a premium for a consultant who actually exists, who has actual lived experience, and who takes real legal and emotional responsibility for their advice.

Your potential clients are already hyper-paranoid; they are looking for any excuse to hit "delete" and assume you are just another automated scam. If your marketing sounds like it was polished by a committee of boring algorithms, you have already lost the sale before the first meeting.

The moment you stop hiding behind generic corporate jargon is the moment you become the only safe harbor in a sea of automated fakes.

But if you are good at what you do and your calendar is still empty, your messaging is likely the thing quietly killing your credibility. You do not have time to wait for the artificial intelligence revolution to settle. You need to fix the leaks in your funnel today. Get my 5-minute marketing fix and identify the exact spots where your message is making paranoid clients hesitate.

👉 Stop losing sales. Download the fix now.

Related Articles:

1. AI Is Changing Consulting Business Models Fast

Now that you know the intelligence explosion is actively replacing white-collar work, you need a plan. This article shows you exactly how to redesign your consulting pricing and packages around your irreplaceable human value instead of trying to compete with a machine on technical output.

2. Why You Should Never Trust An AI Chatbot To Guess The Odds

While the newest models are writing their own code, they still fundamentally misunderstand human risk and probability. Discover why outsourcing your professional judgment to these rapidly improving systems is still a massive, potentially fatal gamble for your service business.

3. Why That Fake AI Video Should Actually Terrify Your Business

Matt Shumer warns that we are in the "February 2020" phase of AI. This post explores how that exact same blissful ignorance is causing massive consumer paranoia, and why your clients are suddenly struggling to trust anything you publish online.

4. Shadow AI Risk Is Growing Fast

If the machines are getting exponentially smarter every four months, you need to know exactly who in your office is using them. Learn how to stop your employees from quietly plugging highly sensitive company data into these new, self-training intelligence models.

5. Massive Microsoft AI Loss: What Small Businesses Should Learn

Even as the technology accelerates at a terrifying pace, blindly throwing money at it can still ruin your company. Discover how the biggest tech players misjudge these tools, and why clear strategy always beats dazzled, fear-based technology adoption.

FAQs:

1. What does it mean that AI is in the "February 2020" phase?

Tech investor Matt Shumer uses this comparison to highlight our collective denial. Just as we ignored a distant virus right before it completely changed our daily lives, professionals are currently ignoring the massive leaps in artificial intelligence that are about to permanently disrupt their industries.

2. Can artificial intelligence actually write its own code?

Yes. When OpenAI released GPT-5.3 Codex in February 2026, they explicitly stated that the model was instrumental in creating itself. The AI was used to debug its own training and manage its own deployment, a process researchers refer to as an intelligence explosion.

3. Will AI replace office jobs and white-collar workers?

Experts believe it will happen very soon. Dario Amodei, the CEO of Anthropic, predicted that artificial intelligence will eliminate fifty percent of entry-level white-collar jobs within one to five years. The technology acts as a general substitute for cognitive work, improving across all industries simultaneously.

4. Why do people think AI is still bad at its job?

Most people base their opinion on the free, outdated versions of these tools. The free versions from a year or two ago hallucinated and made constant errors, but the current paid models used by tech insiders are exponentially more powerful and can complete multi-hour tasks completely independently.

5. How can I protect my career from being automated?

You must start using the best paid tools immediately, pushing them to do actual complex work instead of just asking them simple questions. Furthermore, you must lean into the things that are hardest to replace, such as building deep human relationships, establishing trust, and taking licensed professional accountability for your decisions.