If a job is built around repetition, Mark Cuban says it is already under pressure.
The billionaire investor and TV personality has warned that AI may not wipe out white-collar work all at once, but could quietly reduce the need for companies to hire people for routine tasks. That shift is already showing up in entry-level roles tied to data entry, bookkeeping, customer support, software development, research, finance, and legal paperwork.
The bigger story is not a sudden robot takeover. It is a slower hiring squeeze, where companies use AI to absorb work that once gave junior employees their first rung on the ladder.
1. Entry-level white-collar jobs
Dario Amodei, CEO of Anthropic, warns that automation is rapidly eliminating entry-level roles and that up to half of them will be automated in the next five years.
Roles such as bookkeeping, data entry, and invoice processing are at the forefront of this shift toward automation. Tools like Zapier Central, Notion AI, Brex AI, and UiPath are becoming efficient with handling these roles.
According to NewsNation, Cuban warns that the impact won’t be centered solely on layoffs, but on fewer openings, as companies slow down or skip entry-level roles altogether.
2. Software engineering
AI tools can now generate, debug, and scaffold code that used to justify hiring for entry-level roles. That’s shifting demand towards skilled software engineers with knowledge of systems design to guide AI, not just write routine code.
Agentic tools like Lovable, Cursor, and Replit are accelerating this shift with fewer junior developers and a leaner team.
3. Customer service roles
Businesses are now integrating custom-built AI chatbots on their websites and using AI voice agents that resolve most tier-one support tickets without human involvement, improving response time and customer satisfaction.
That shift leaves call center operatives and support desk employees competing with tools like Zendesk, Connect Customers, and Salesforce Einstein Service Agent.
4. Data analysis and research
Data analysis is quickly moving from completely human-led to AI-led, and in some cases, AI-driven.
AI tools can now summarize reports, surface insights, and answer business questions in seconds. The analysts that survives this are the ones who know the right prompt to ask, catch when the tool hallucinates, and can turn an AI output into a decision a business executive can act on.
With tools like Julius AI, Tableau GPT, and Perplexity AI Enterprise shaping this industry, the need for manual analysis is shrinking.
5. Finance and law work
While lawyers and finance experts will remain, a significant part of their work will become AI’s job to do. Tasks like document review, contract analysis, compliance checks, and paperwork are now being automated. Critical functions such as strategy, negotiation, and judgment will most likely persist.
Some of the tools already handling this work include: CoCounsel, Booke AI, and Blue J.
Bottom line: Cuban’s recommendation
Despite the advances AI has made, Cuban stresses that humans will not be completely replaced but will experience a shift in the roles they require. That is because AI lacks situational awareness and is prone to hallucinating.
The role of humans now moves from handling mundane tasks to piloting AI systems. To do this, he advised employees to stack relevant AI skills that enable them to solve problems with AI tools, rather than avoiding AI or letting it “do the thinking.”
The divide here won’t be between machines and humans, but between those who can and are willing to use it where relevant and those who’ve refused to evolve.
For those looking to adapt, not just react, here are 10 AI jobs you can land without a computer science degree, from prompt engineering to AI training roles.