Every conversation about AI and jobs gets stuck in the same loop. AI will take X million jobs by Y year. This profession is at risk. That one is doomed. The narrative is so loud that it drowns out the other half of the story.
Here’s the other half: AI is also creating entire categories of work that did not exist a few years ago. Not future roles. Not speculation. Actual jobs that people are being hired into right now, with real salaries, real career paths, and real demand that exceeds supply.
I’ve written before about why I don’t buy the doom narrative and why the question “which jobs are safe?” is the wrong question. This post is the proof. These seven roles either didn’t exist three years ago or existed only as obscure niches. Today, they’re some of the fastest-growing careers in the economy.
1. Prompt Engineer
Pay range: $101,000 to $173,000
Three years ago, this job title was a joke. Today, prompt engineering is a legitimate discipline that involves designing structured, repeatable ways of communicating with AI systems to get reliable, high-quality outputs. The best prompt engineers combine clear writing, systems thinking, and an understanding of how language models behave.
Companies are paying for it because unreliable AI outputs are expensive. A prompt engineer who reduces hallucinations by 40% pays for their own salary many times over.
2. AI Ethicist / Responsible AI Lead
Pay range: $93,000 to $137,000
As AI gets deployed in healthcare, finance, hiring, and criminal justice, someone has to make sure it isn’t causing harm. AI ethicists audit models for bias, design governance frameworks, and serve as the bridge between engineers and the real-world consequences of what gets built.
This role barely existed five years ago. It’s now one of the fastest-growing positions in tech, with projected growth of 29% through 2032.
3. AI Trainer (RLHF Specialist)
Pay range: entry-level to $80,000+
Every major AI system you’ve used has been shaped by human trainers. They evaluate outputs, flag problems, and provide the feedback that teaches models what “good” looks like. The technical term is Reinforcement Learning from Human Feedback, and it’s one of the most important jobs in AI development.
What’s interesting is how accessible this role is. You don’t need a computer science degree. You need judgment, clear thinking, and a willingness to iterate. This is a real entry point into AI work for people coming from teaching, writing, or editing backgrounds.
4. AI Agent Supervisor
Pay range: varies widely, $60,000 to $250,000+ depending on seniority and scope
This is the newest role on this list, and it’s exploding. AI agents are software that takes actions on its own — researching, writing, sending emails, making purchases. Someone has to design these agent systems, monitor their behavior, and step in when they go off the rails.
Entire platforms are emerging that let individual operators supervise fleets of AI agents to run research, marketing, and operational workflows. The skill is less about coding and more about knowing what good work looks like and catching when the AI doesn’t.
5. AI Product Manager
Pay range: $150,000 to $300,000+
Traditional product management is about understanding users and shipping features. AI product management adds a harder problem — how do you ship a product when the underlying technology is probabilistic, constantly improving, and sometimes gets things wrong?
Every major company is hiring these people right now. The best ones are unicorns who blend product intuition, technical literacy, and the ability to translate between engineering teams and business stakeholders.
6. AI Red Teamer / Model Evaluator
Pay range: $120,000 to $250,000+
Red teamers try to break AI systems before bad actors do. They probe for vulnerabilities, jailbreaks, harmful outputs, and edge cases that developers didn’t anticipate. It’s part hacker, part ethicist, part creative writer.
This role emerged out of the AI safety conversation and has quickly become standard at every major AI lab. Demand is high and the work is genuinely interesting — you get paid to think adversarially about how technology could fail.
7. AI-Native Micropreneur
Pay range: highly variable, from zero to seven figures
This one is harder to categorize because it’s not a job you apply for. It’s a new way of working that AI has made possible. Individual operators are using AI agents to run businesses that would have required teams of ten people a few years ago — writing content, managing customer service, running marketing campaigns, analyzing data.
The skill set is entrepreneurial, not technical. You need to spot opportunities, experiment quickly, know your strengths, and stay adaptive as tools evolve. This is probably the most important emerging category because it’s a template for how a lot of work will look going forward.
What All Seven Have in Common
Look closely at this list and you’ll notice something. None of these jobs are about competing with AI. They’re all about working alongside it — shaping it, directing it, evaluating it, supervising it, building with it.
The skills that make someone good at any of these roles are the same skills I keep coming back to on this site. Judgment. Creativity. Clear thinking. Adaptability. Ethical reasoning. The willingness to learn new tools every few months without losing your grounding in what actually matters.
Three years ago, this list didn’t exist. Five years from now, there will be another list with different roles nobody has thought of yet. The winning move isn’t picking the right job from the current list. It’s getting good at spotting the pattern so you can identify the next wave before everyone else.
The future of work isn’t something being done to you. It’s something you can actually participate in building — if you’re paying attention.
This post was co-written by Jason Batten and Claude Cowork.