AI is no longer just a threat to factory and transportation jobs. As AI grows and remote work becomes more common, entry-level office workers may also be at risk.
Mark Cuban has warned that five job categories could be especially vulnerable as companies use more AI. Rather than disappearing overnight, these jobs may slowly become harder to find, making it more difficult for young workers to gain experience and start their careers.
The common thread: "binary" tasks
Before getting to the specific categories, it helps to understand the logic Cuban uses to identify which jobs are most exposed.
Cuban has pointed to what he calls "binary" tasks as the most vulnerable: structured, repetitive work with clear inputs and outputs that AI systems can process faster and at a larger scale than humans. This includes data entry, document review, basic research compilation, and routine compliance checks. These are tasks where the answer is either right or wrong, the process is consistent, and the result is predictable. That profile is exactly what current AI tools are built to handle.
Cuban has also noted that "humans have a far greater capacity to know the outcomes of their actions," which is his way of saying that AI still lacks genuine judgment, contextual awareness, and the ability to anticipate consequences in complex real-world situations. The tasks that require those capacities remain more protected; the tasks that do not are the ones he is worried about.
Entry-level white-collar roles
Cuban identified entry-level white-collar roles broadly as among the most exposed, specifically data entry and bookkeeping as examples of the "binary" tasks AI is already absorbing. These are roles where a recent graduate or career starter has historically spent their first year or two building foundational skills. If AI handles the task before the hire is ever made, that foundational experience never happens.
The concern is not just the job, but what the job leads to. Entry-level roles are where workers learn processes, build professional habits, and develop the judgment that comes from doing work and seeing the results.
Cuban has warned that AI could lead to fewer openings and slower hiring in these categories without eliminating the category outright, making the career path harder to start even if experienced positions still exist.
Software development (entry-level)
Cuban has acknowledged that AI-assisted coding tools are widely used and expects them to reduce the value of routine programming tasks, particularly the kind of boilerplate and documentation work that junior developers typically handle.
Higher-level responsibilities like system design, architecture decisions, and complex problem-solving are less exposed because they require exactly the contextual judgment AI does not yet have.
Customer service
AI chatbots and voice systems are already handling many basic customer questions, and that trend is likely to continue. Human workers are increasingly needed for more complicated situations, such as difficult complaints or conversations that require empathy and good judgment.
That means entry-level customer service jobs may become harder to find. Many simple tasks once handled by new employees can now be done by AI, while the jobs that remain often require more experience and stronger people skills.
Research and data analysis
Cuban has noted that AI tools can now summarize datasets, generate reports, and identify trends, overlapping with work traditionally performed by junior analysts.
The shift, in his view, moves the value away from producing analysis and toward interpreting it: knowing what questions to ask, what context matters, and what the results actually mean for a specific business situation.
That interpretive capacity requires experience and domain knowledge that entry-level workers have not yet had time to build. The result is a category where the work that once justified hiring a junior analyst is increasingly automated, but the skills needed to do the remaining work are not yet those a new hire possesses.
Finance and legal support
Routine work in finance and legal departments, including document review, compliance checks, and basic accounting functions, is particularly vulnerable according to Cuban, because it fits the binary task profile precisely. A document review that once required a paralegal or junior associate can now be completed by AI tools that scan for specific language, flag exceptions, and produce summaries far faster than a human reviewer.
Cuban noted that experienced professionals in these fields may still be in demand because the judgment calls at higher levels remain complex and consequential.
What Cuban says workers should actually do
Despite the warning, Cuban does not predict a widespread collapse in employment. He has consistently framed the current moment as a period of disruption similar to past technological shifts, such as the rise of personal computers, when some roles declined while new ones emerged.
His consistent advice is to learn to use AI tools rather than avoid them. Workers who understand how to direct, verify, and apply AI output are positioned on the productive side of the disruption.
The human capacities he emphasizes as durable are judgment, creativity, communication, and the ability to solve problems that do not have structured answers. In his view, "There's going to be two types of companies. Those who are great at AI, and everyone else," and the workers inside those companies face the same divide.
Bottom line
Cuban isn't saying these careers will disappear. Instead, AI may eliminate many entry-level roles that traditionally help workers gain experience and move up the career ladder.
If you're starting your career or looking to earn extra money in one of these fields, learning to use AI could give you an edge. Workers who know how to prompt, evaluate, and apply AI tools may be better positioned as employers increasingly automate routine tasks.
Add Us On Google