The Hidden AI Job Boom No One's Talking About
The truth about AI and employment is nothing like what 'experts' are telling you.
Most people think AI is coming for their jobs.
They’re not wrong. But they’re not right either.
Because while AI will replace some jobs, it’s already creating entirely new ones.
I know—because I build Agentic solutions for a living.
And here’s what the folks discussing this subject won’t tell you:
AI in the enterprise doesn’t eliminate the need for humans. It just changes what “human work” looks like.
Want to know more?
Let’s dive in.
The AI Employment Deception
This week, the "Diary of a CEO" podcast wandered onto the territory of AI and job displacement.
Their discussion quickly veered into alarming predictions about widespread unemployment:
"If your job is as routine as it comes, your job is gone in the next couple years."
"If your job is to produce some kind of artifact that's like probably text or images, that job is is at risk."
"Women are disproportionately affected by automation, with about 80% of working women in an at-risk job compared to just over 50% of men."
These statements come from influencers and theoretical thinkers—not people who have actually implemented enterprise AI at scale or built detailed cost models for deployment.
They're speaking from conceptual understanding, not practical experience.
Notice what's missing from these pronouncements: any mention of new roles created, implementation costs, regulatory requirements, or the human infrastructure needed to make AI systems functional in real-world settings.
When someone from Diary of a CEO says "accountants are gone," they're not accounting for the fact that enterprise AI systems require human expertise to function—they've just shifted where that expertise is applied.
The truth?
Agentic AI systems don't just delete jobs—they transform them and create entirely new roles that require human intelligence, oversight, and domain expertise. And critically, these roles often pay better than the positions they evolved from.
The AI Transformation Reality
From where I’m sat, what's really happening is a two-stage transformation.
First, AI agents unlock latent human capacity by automating routine tasks; then, this shift creates demand for new specialised roles to support, enhance, and govern these AI systems.
The result isn't mass unemployment—it's a workforce evolution where humans move up the value chain.
Let me show you exactly what this looks like in practice.
At Templonix, we prototype agent systems for enterprise.
Think: document automation, customer ops, sales tools—but with built-in memory and reasoning.
In every project we work on, here's what actually happens:
AI doesn't erase all the jobs. It transforms them. And it creates new ones.
We call this the latent capacity unlock – freeing smart people from tedious work and repositioning their effort where it adds real value.
Take our recent case with a financial services company: Before implementing AI agents, their 76-person sales team spent 10.5 hours per month each on pre-meeting research—that's 800 total hours monthly, equivalent to 5 full-time employees and £643,200 in direct annual costs.
After implementing an AI Sales Support Agent, they reduced research costs by 98.8% (from £268 to £3 per research task) and enabled the sales team to spend more time on high-value client interactions.
The result?
A 2% sales increase across the team, generating £395,000 in additional annual revenue with a 1000% return on investment and a one-year payback period.
But here's what the "experts" miss entirely: this implementation didn't eliminate sales jobs—it created entirely new roles to support their new AI powered business.
The 6 New Roles AI Is Creating Right Now
Before we dive into the roles themselves, take a look at the heatmap below. It gives you a sense of where these opportunities sit on the dimensions that matter most:
How hard they are to transition into?
What kind of pay bump you can expect?
Some of these are relatively easy jumps for professionals already in the field.
Others require a steeper learning curve but can open the door to six-figure salaries and strategic influence.
1. The Human Supervisor: AI's Essential Safety Net
Role Status: Emerging.
Far from eliminating humans, many AI deployments (and especially agents) create a need for human supervisors to oversee and validate AI outputs. These professionals act as a safety net, monitoring AI agents in production and handling exceptions the AI can't resolve. This pattern is common across industries adopting AI for critical workflows.
Organisations implementing AI in high-stakes processes (finance, healthcare, content moderation, etc.) are embedding human supervisors. Job postings may not always use the title "Human Supervisor," but roles like "AI Operations Associate", "AI Quality Analyst", or "Human-in-the-Loop Specialist" reflect this function.
Because these roles are often filled by experienced domain experts transitioning from the tasks AI now handles, compensation remains relatively strong. In one case, Rocket Mortgage has automated 70% of document review and kept its senior loan processors on as AI supervisors at only ~15% lower pay than their previous analyst jobs (while eliminating much of the drudgery).
Your Transition Path: This is one of the most accessible entry points into AI-adjacent roles. If you're currently in a position being augmented by AI, your deep domain knowledge is your greatest asset.
Companies value people who understand the nuances, edge cases, and quality standards of the process being automated.
The transition typically requires supplementing your existing expertise with basic AI literacy—understanding how models work, their limitations, and how to spot potential issues. Many employers offer this training in-house as they deploy AI systems, making this an ideal transition for professionals who want to stay in their industry while evolving their role.
While this is the least technically challenging role to move into, it’s also the one where the pay performs the lowest.
2. The Prompt Engineer: Getting Machines to Think
Role Status: Emerging (High Demand).
The Prompt Engineer is often cited as the quintessential new AI-era job. These specialists craft the textual instructions or questions ("prompts") that guide large language models and other generative AI systems to produce useful output.
AI-forward companies – from startups like OpenAI and Anthropic to tech giants integrating AI – have hired prompt engineers. Notably, Anthropic made headlines with a job listing for a Prompt Engineer & Librarian with an eye-popping salary range of $175,000–$335,000.
Early demand has driven salaries very high for top prompt engineers. These figures grabbed headlines (e.g. "AI prompt engineer roles offering over $300k, including one at Anthropic"). That said, not every prompt engineer makes six figures – outside of elite tech firms, salaries can be more modest.
Your Transition Path: Content creators, technical writers, and QA specialists are perfectly positioned for this role because writing good prompts is all about structure and clarity. The key transition skill is learning to "speak AI" – understanding how language models interpret instructions and their limitations.
Knowledge of NLP and basic programming or scripting certainly helps, but since the job revolves around trial-and-error experimentation, it is possible to do this role without hardcore technical skills and direct experience of working with developers.
The demand for this skill exceeds the supply, especially for those who can demonstrate a systematic approach to prompt development and testing. If you have strong writing skills and logical thinking, this might be your fastest path into the AI economy.
3. The Vector DB Engineer: The Memory Architects
Role Status: Emerging and in Demand.
As companies incorporate AI that relies on vast unstructured data (think GPT-based assistants that need facts from your documents), vector databases have become critical infrastructure. A Vector DB Engineer designs and maintains these new databases optimised for storing embeddings (numerical vectors representing data). In plain terms, they ensure the AI's "memory" or knowledge base is efficiently searchable.
Initially, specialised AI infrastructure companies (like Pinecone, Weaviate, Qdrant) were obvious employers. Now, mainstream tech firms and enterprises are seeking these skills too. Even companies like Tesla have job posts asking for "hands-on experience with Retrieval-Augmented Generation (RAG) and vector databases" as part of AI engineering roles.
Because it's so new, talent is scarce – which translates to premium pay. Anecdotally, some companies are offering 40–60% higher salaries for engineers with vector database skills compared to standard database roles. A senior data engineer might earn $100K, but a comparable vector DB engineer could command $140K–$160K.
All signs point to the Vector DB Engineer being a legitimately new role born of AI's need for better knowledge retrieval, with a strong career runway ahead.
Your Transition Path: Traditional database administrators and data engineers have a natural pathway into this specialty. The transition requires learning about embedding models, similarity search algorithms, and the specific requirements of vector databases.
While this role has a steeper technical learning curve than some others, the premium compensation reflects that investment.
Start by familiarising yourself with vector search concepts, then gain hands-on experience with open-source vector databases like Chroma or Milvus. Many organisations are willing to train promising database professionals internally because finding external talent is so difficult.
4. The Agent Architect: The AI System Designers
Role Status: Emerging (Highly Specialised, High-Paying).
An Agent Architect designs complex AI systems composed of multiple interacting agents or models. As companies push beyond using a single chatbot to deploying swarms of AI agents that can perform different tasks collaboratively (e.g. one AI that plans, another that executes, another that verifies), someone has to design that orchestration.
Forward-looking tech organisations and cloud providers are investing in this area. For example, Amazon Web Services recently listed an opening for a "Principal Generative AI Solutions Architect (Agentic systems)" whose job is to "architect multi-agent ecosystems where specialised AI agents collaborate through sophisticated orchestration frameworks to solve complex enterprise challenges."
This role often commands top-tier compensation, on par with senior software architects or higher. In the AWS job posting mentioned, the base pay range for that principal architect was $164,500 up to $284,300 – and that's before stock or bonuses. In the UK/EU context, experienced contractors in this niche reportedly bill over £1,500 per day for their services.
This role requires a “systems thinking” mindset and broad technical mastery. Most Agent Architects come from senior software engineering or architecture backgrounds. Key skills include designing distributed systems, API integration, and knowledge of how different AI components (LLMs, tools, memory stores, etc.) fit together.
They must also be intimately familiar with the capabilities and limitations of AI models (so they know which agent should handle which task, and what failure modes to expect). Experience with frameworks for building agents (like LangChain or similar orchestration libraries) is often expected. In short, an Agent Architect is part software architect, part AI researcher.
Your Transition Path: Software architects and senior developers have the most direct path into this role. The essential transition requires developing a deep understanding of AI model capabilities and limitations, then applying systems thinking to how these models can work together.
Start by building proof-of-concept projects that integrate multiple AI components and demonstrate your understanding of orchestration patterns. Practical experience with LangChain, LlamaIndex, or similar frameworks is valuable, as is familiarity with the challenges of managing AI memory systems and handling the interaction between agents.
While challenging, this transition offers some of the highest compensation potential in the AI space.
5. The AI Governance Officer: The Guardrail Builders
Role Status: Emerging to Established (fast becoming standard in big firms).
As companies deploy AI at scale, they face questions of ethics, compliance, and risk. AI Governance Officers (or Responsible AI Leads) are charged with creating the policies and frameworks to ensure AI is used ethically and in line with regulations.
A broad range of organisations are hiring, notably: financial services (banks, insurance), healthcare companies, pharmaceuticals, tech giants, and even governments. For example, Novo Nordisk (a global pharma) recently hired an AI Governance Lead to ensure patient safety and regulatory compliance in its AI initiatives.
These roles often sit in higher corporate tiers (risk management or legal/compliance departments), so they tend to pay accordingly. A mid-level AI governance or ethics specialist might earn in the low-to-mid six figures in the US.
These professionals blend legal/regulatory expertise, ethical reasoning, and technical understanding. Many come from compliance, legal, or risk management backgrounds and supplement with AI knowledge.
Your Transition Path: Legal professionals, compliance officers, and risk managers are ideally positioned to move into AI governance. The key transition skill is bridging regulatory understanding with sufficient AI technical knowledge to develop practical governance frameworks. This doesn't require becoming a data scientist—it requires understanding AI risks, limitations, and ethical implications well enough to create intersection of technology, policy, and ethics.
6. The AI Risk Advisor: The Strategic Guarders
Role Status: Emerging (Especially in Finance & Consulting).
An AI Risk Advisor focuses on identifying and mitigating the risks that arise from deploying AI solutions. This role is closely related to AI governance but leans more into risk assessment frameworks, security, and business continuity.
The financial sector is a prime employer – banks, insurance companies, and fintechs are subject to strict model risk management guidelines and thus hiring AI risk experts. Consulting firms are also hiring AI risk advisors to offer services to clients concerned about deploying AI.
These roles typically pay on par with other risk management and IT audit roles, often six figures. A mid-level AI Risk Specialist might see salaries around $120K–$150K in the US, based on postings.
An AI Risk Advisor needs a strong grasp of risk assessment methodologies (identifying likelihood and impact of various AI failure scenarios) and familiarity with AI "failure modes" – e.g. knowing how models can go wrong, from biased outputs to cybersecurity vulnerabilities.
Your Transition Path: Information security professionals, business continuity planners, and risk analysts have a natural pathway into this specialty. The transition requires learning AI-specific risk frameworks (like NIST's AI Risk Management Framework) and developing a deep understanding of AI failure modes.
Start by mapping your existing risk methodology experience to AI-specific challenges like data privacy, model bias, and robustness.
Many financial institutions are now developing their own training programs for this role, recognising that finding external talent with both AI and risk expertise is challenging. This transition path is ideal for professionals who enjoy systematic risk identification and mitigation and want to apply those skills to emerging technologies.
AI’s Impact on Jobs: The Reality
Yes, AI is changing the job market – but not in the catastrophic way some headlines suggest.
The "Fully Autonomous" Fantasy
The phrase "fully autonomous AI" with no human oversight is what I call a "Kentucky fried mouse"—it looks appetizing on the surface, but bite into it and you'll get a nasty surprise.
Here's why:
Regulatory requirements mandate human accountability in most industries
Edge cases consistently require human judgment
Business processes evolve, requiring constant retraining and supervision
Liability concerns make fully autonomous AI legally problematic
Trust issues with stakeholders who demand human oversight
Augmentation, Not Replacement
Accountants, lawyers, designers – they're not being replaced. They're being augmented. AI is handling the grunt work, leaving professionals more time to focus on strategy, advisory, and creative thinking. Many of these roles are also spawning new specialisations like prompt engineering and AI governance.
And the idea of a fully autonomous AI workforce? That's fantasy – for now.
Every successful AI deployment I've seen still relies on human judgment to supervise, course-correct, and validate output. In fact, most firms who rushed to replace humans later admitted it hurt performance.
The future isn't jobless. It's different.
If you're willing to evolve, it's also full of opportunity.
How to Position Yourself for the AI Transition
If you're worried about AI impact on your career, here are three immediate steps I recommend:
Identify your transferable domain knowledge - Your industry expertise remains valuable and can't be replicated by AI.
Select your transition target - Based on your current skills and interests, which of the six roles aligns best? The transition difficulty varies significantly.
Start skill-building now - Online courses in prompt engineering, vector databases, or AI governance can give you a head start before your company begins its transformation.
The key insight from our data: those who proactively adapt their skills survive and thrive during AI transition. Those who wait for disruption struggle.
The Bottom Line
You don’t need to fear AI taking your job. You need to fear ignoring the opportunity to evolve into the roles AI is creating.
The real transformation isn’t about replacing people. It’s about unlocking capacity and scaling expertise.
If you’re ready to adapt, the next 5 years could be the best career window of your life.
And if you’re still listening to hype merchants yelling "AI is coming for your job!" — maybe stop taking career advice from clickbait.
Until the next one,
Chris.
🧰 Whenever you're ready, I might be able to help you.
The Agent Architect's Toolkit gives you the practical expertise that transforms casual AI interest into career-advancing competence. Enterprise AI knowledge that makes you the expert in the room - for less than a gym membership.
This article attempts to paint a hopeful vision of AI's impact on work, but in doing so, it reproduces many of the same blind spots it critiques. It’s brand malpractice for someone in the business of selling “Agentic” systems to admit that their products will reduce headcount—and that bias shapes every frame in this piece. It’s not dishonest, but it’s structurally compromised.
The author lists six “new” job types—Human Supervisor, Prompt Engineer, Vector DB Engineer, Agent Architect, AI Governance Officer, and AI Risk Advisor—but doesn’t acknowledge that many of these roles aggregate the labor previously performed by multiple specialists. A single Agent Architect might replace an entire team of developers and product managers. One Prompt Engineer might do the work of a strategist, writer, editor, and QA lead. These aren’t apples-to-apples substitutions—they’re compression artifacts in a collapsing labor market.
And that collapse is structural.
This is a fixed point on a larger trajectory—the shift from Software-as-a-Service to Employee-as-a-Service, in which intelligent systems are trained on human labor, and then redeployed to displace or disaggregate it. At the same time, we’re seeing the VP-of-AI to AI-VP pipeline emerge: automating leadership itself. As AI agents gain reasoning capacity, they’re not just support—they’re replacement candidates for mid-tier and senior-level roles.
Why is this happening? Because the middle class was never purely meritocratic—it was a functional pseudo-UBI, sustained by bureaucratic inefficiencies and gatekept credentials. As soon as those inefficiencies become optimizable, they’re stripped out. That’s not a revolution. That’s a system metabolizing its own redundancy.
The article proudly shares stats like a 98.8% cost reduction for sales research and £395,000 in extra revenue—but doesn’t ask where those five “unlocked” full-time employees go next. Or how many waves of layoffs we’ve already seen in tech, media, and operations. Or how new “AI-adjacent” jobs, while real, aren’t proliferating at the scale necessary to absorb the displaced.
We don’t need total job loss to have catastrophic outcomes.
All it takes is enough friction in workforce reabsorption, enough hollowness in career paths, and enough centralization of capability—and we get a kind of economic organ failure.
The hidden job boom isn’t a lie. It’s just not the whole story.
The AI slop writing style reads like '70s knit polyester feels.