.png){kind=logo}
Generative AI is raising productivity while changing how companies structure roles, particularly for junior and entry-level workers.
Freelance markets are shifting faster, with declining demand for routine writing, coding, translation, and design work.
Workers who combine AI with specialist expertise, judgment, quality control, and client relationships are better positioned to retain their value.
Generative AI is not just automating tasks. It is changing what it costs to produce routine knowledge work. Writing a first draft, building a template, or drafting basic code used to take hours of human time. A chatbot now does it in seconds. That shift is starting to move wages, hiring plans, and freelance pricing at once.
The scale is already global. A 2025 study from the International Labour Organization found that one in four workers worldwide sit in jobs with some exposure to generative AI. Only 3.3% fall into the highest exposure group. Most jobs mix tasks AI can handle with tasks that still need a person, which is why the ILO expects transformation, not mass replacement, to be the more common path.
The clearest data point comes from a 2023 NBER study led by economist Erik Brynjolfsson. His team tracked 5,179 customer support agents after a software company gave them an AI assistant. Output, measured as issues solved per hour, rose by 14% on average. Newer and lower-skilled agents gained the most, improving by about 34%. Experienced agents barely moved.
The pattern suggests AI works less like a multiplier for everyone and more like a floor raiser. It lifts the weakest performers close to the level of the best ones. Similar effects show up in early research on coding and writing tasks, though the size of the gain shifts with the setting.
The sharpest risk sits with junior workers. AI now handles many of the small tasks that once taught new hires their trade. Drafting a contract, writing a first pass of code, or producing early marketing copy used to build skill over months.
If AI absorbs that layer of work, the real question turns to where a junior worker gains the practice needed to grow into a senior one. That is a structural problem for career paths, not a simple case of fewer openings.
Freelance platforms show the change in near real time, since pricing and job counts move fast. A study covering close to two million freelance postings across 61 countries found that within eight months of ChatGPT's launch, writing gigs fell by about 30%, software postings by about 21% and design work by about 17%, with no sign of a rebound.
A separate paper in the Journal of Economic Behavior and Organization adds nuance. Demand for skills AI can replace, such as writing and translation, dropped 20 to 50%. Demand for skills that pair well with AI, such as machine learning support, moved up. Total freelance demand on the platform studied did not fall overall.
The mix simply shifted. Research covered by Wharton also found that freelancers active before ChatGPT cut their job applications by 51 to 62% in the months that followed, as rivals flooded the exposed categories.
Here is the part often missed. Higher output from AI does not turn into higher income on its own. Three things can happen instead. Prices can fall as more people compete for the same work. A worker can take on more clients and offset lower per-project rates with volume, or an employer can simply need fewer people to hit the same target.
Which path wins depends on demand growth, how easy the work is to copy, and how much leverage workers still hold. In writing and other easily copied categories, price pressure has so far won out.
Also Read: Why Mass Applying for Jobs with AI May be Hurting Your Chances
Deep expertise in a narrow field still counts for a lot. So does the skill of checking AI output for errors, framing a problem clearly enough for a tool to solve it well, and owning the final result rather than just the first draft.
Client relationships and plain communication also carry more weight now that raw production is cheap. The next stage of this shift will likely reward people who direct AI well over people, who compete against it on raw speed.
Also Read: AI Job Cuts, Burnout Rise as Workplace Stress Grow
The workplace of the next few years will probably split along a new line. On one side sits work that AI can copy at almost no cost. On the other side is work that stays valuable because a person stands behind the outcome. Firms that build training paths around this split, rather than just cutting junior headcount, are likely to keep the deepest bench of talent as the tools keep improving.
1. How is generative AI affecting employment?
Generative AI is changing job roles by automating routine tasks and increasing the need for human oversight, judgment, problem-solving, and accountability.
2. Which workers are most affected by AI?
Entry-level and junior workers face greater pressure because AI can perform many routine tasks that traditionally helped employees build experience and skills.
3. How is AI affecting freelance work?
AI has reduced demand for some freelance services, particularly routine writing, translation, coding, and design work that can be produced quickly with AI tools.
4. Can AI productivity gains lead to higher wages?
Not necessarily. Higher productivity can increase output, but wages may remain flat or decline when businesses can produce the same amount of work with fewer workers.
5. Which skills will remain valuable as AI adoption grows?
Specialist expertise, critical judgment, quality control, client communication, problem framing, and responsibility for final outcomes are likely to remain valuable.
