If you only read headlines, you might believe that artificial intelligence has gutted the white-collar workforce. The reality, three years into the generative AI boom, is more nuanced and considerably less apocalyptic. Some categories of work have genuinely shrunk. Others have grown in ways nobody predicted. Most have simply changed.

Customer support is the clearest case of contraction. Tier-one support roles, the kind that follow scripts and answer routine questions, have declined by roughly thirty percent in the United States since early 2023. Companies have shifted that volume to AI agents, and the human roles that remain are now more senior, handling escalations and edge cases that automated systems struggle with. Pay for the remaining positions has actually risen.

Software engineering is more interesting. Despite predictions that AI would replace junior developers, the data shows the opposite trend at the team level. Engineering teams now ship more code than they did three years ago, and demand for engineers who can effectively orchestrate AI tools has risen sharply. What has changed is the nature of the work. Engineers spend less time writing boilerplate and more time reviewing, debugging, and architecting. The role looks different. It has not disappeared.

Marketing, copywriting, and content production tell a similar story. The bottom of the market, generic blog posts and templated social media copy, has been hollowed out. Strategic roles, brand work, and senior creative direction have actually become more valuable. Companies have discovered that AI can produce infinite mediocre content cheaply, which has paradoxically increased the premium on genuinely good work.

Healthcare, education, skilled trades, and most in-person service jobs remain largely untouched. These sectors are growing in absolute terms, driven by demographic trends that AI cannot reverse. If anything, the labor shortage in nursing, elder care, and skilled construction has gotten worse.

For job seekers, the practical implication is straightforward. Roles that involve judgment, relationships, physical presence, or genuine creativity have become more valuable. Roles that involve high-volume, low-judgment information processing have become riskier. The middle of the market is being squeezed from both directions.

The workers who are thriving in 2026 are not the ones who avoided AI. They are the ones who learned to use it as a collaborator. If your job involves a computer, the most important career investment you can make this year is probably learning how to use these tools well. The rest will follow.