Artificial intelligence is changing the job market in a way that is easy to miss if the conversation focuses only on replacement. A review of 47,000 job listings points to engineering roles being created as AI becomes part of the work itself, while employment for software developers is projected to increase by about 175,000 between 2025 and 2035.
Those numbers tell two connected stories. AI is influencing the tasks inside many occupations, but that influence does not automatically mean fewer jobs. At the same time, software development is expected to add a large number of positions over the next decade, showing that demand for people who build and work with software remains part of the employment picture.
What 47,000 listings say about AI work
The 47,000 job listings offer a look at the engineering roles that AI is creating. The figure matters because it focuses attention on hiring activity, not only on predictions about what AI might do in the future. Job listings show where employers are seeking workers as AI becomes connected to engineering and software development.
The listings also give the broader debate a more useful frame. AI can assist with tasks in an occupation, yet the occupation can still see employment growth. The projected increase of about 175,000 software developer jobs between 2025 and 2035 fits that picture: AI can change how software work is done while demand for software developers continues to rise.
That does not describe every occupation or guarantee the same outcome across the job market. It does show why job creation and AI exposure need to be considered together. One measure tracks the tasks AI can affect; another tracks the number of jobs expected to exist.
AI exposure does not tell the whole employment story
The Bureau of Labor Statistics classified 206 of 831 jobs as very high in AI exposure. Its description explains what that label means: “‘Very high’ relative AI exposure generally indicates that, compared to other occupations, a larger fraction of an occupation’s tasks can be completed or assisted by AI technology, and that LLMs have been observed performing some of the occupation’s tasks.”
That classification describes the share and type of tasks connected to AI. It does not say that all jobs in the category will disappear, nor does it measure whether employment will grow or shrink. The software developer projection helps show why those questions must remain separate.
There is also a wage point in the figures. Over 100 jobs classified as very high in AI exposure earned a median annual wage of at least $75,000 last year. High exposure, then, is not limited to jobs with low pay. It includes occupations where the median annual wage reached at least that amount.
A more complicated picture for workers
Taken together, the figures point to a job market being reshaped along several lines at once. AI exposure identifies occupations where a larger fraction of tasks can be completed or assisted by AI. Job listings show engineering roles being created around that change. Employment projections show software developer jobs increasing by about 175,000 between 2025 and 2035.
The wage data adds another layer. More than 100 very high AI exposure jobs reached a median annual wage of at least $75,000 last year, so exposure alone cannot serve as a simple measure of job quality or pay. A job can have many tasks affected by AI and still belong to a group with a median wage at or above that level.
For workers, the clearest message is that AI is tied to both task changes and new hiring. The 47,000 listings represent engineering opportunities connected to AI, while the 206 of 831 classification shows how many jobs fall into the very high exposure group. Neither number replaces the other.
The projected 175,000 increase in software developer employment gives the story its strongest forward-looking figure. AI is changing the work software developers do, but the employment projection points to more software developer jobs between 2025 and 2035. That combination makes the future of engineering work more complicated than a simple count of jobs lost to technology.
Based on
