AI and skilled trades: why plumbers, electricians, and HVAC technicians are less exposed to automation
Jobs that require physical presence and real-world diagnostics are proving to be the most resilient careers against artificial intelligence in today’s economy
Automation has hit the job market hard, and a lot of people are still trying to figure out what it means for the future of their careers. Increasingly sophisticated AI models are taking over tasks that, not long ago, seemed exclusively human — from writing reports to analyzing complex data in seconds. But while entire offices are rethinking their processes, one group of workers seems to be sleeping just fine at night, and for good reason.
A study from Anthropic, one of the biggest artificial intelligence companies in the world, delivered a finding that flipped the narrative upside down. Jobs that require physical presence, practical reasoning, and real-time diagnostics — like plumbers, electricians, and HVAC technicians — are among the least exposed to AI-driven automation. This is not a coincidence, and it is not luck either.
There is a very concrete technical reason behind this, and understanding it might change the way you think about the value of hands-on skills in today’s economy. Let’s dig into what the data shows and why the skilled trades are becoming one of the safest territories in the middle of the artificial intelligence revolution 🔧⚡
What Anthropic’s research revealed about jobs and automation
Anthropic analyzed millions of real conversations with its AI model, Claude, to understand which professional areas people rely on artificial intelligence the most to get work done. The results were, to say the least, surprising for anyone who believed blue-collar workers would be the first to feel the impact of automation.
In practice, the study showed that the heaviest AI usage is concentrated in cognitive and digital activities — things like programming, writing, data analysis, customer service, and remote tech support. These are exactly the roles that keep showing up in reports about workforce replacement by large-scale language models, commonly known as large language models.
The research found that skilled trade professions — the ones that depend on physical diagnostics, tool handling, and decision-making in unpredictable environments — showed very low levels of exposure to AI automation. A plumber who needs to assess a hidden pipe leak behind a wall, for example, relies on a combination of sensory experience, spatial reasoning, and real-time adaptation that no language model can replicate remotely.
The same goes for an electrician who needs to trace a damaged circuit in an older building, where precise documentation of the original wiring doesn’t always exist. Every situation is unique, every environment presents different variables, and the professional has to process all of that on the spot, with their own hands and senses as their primary diagnostic tools.
Receive the best innovation content in your email.
All the news, tips, trends, and resources you're looking for, delivered to your inbox.
By subscribing to the newsletter, you agree to receive communications from Método Viral. We are committed to always protecting and respecting your privacy.
Another key point from Anthropic’s research is that even when skilled trade professionals use digital tools or AI assistants, the role of artificial intelligence stays limited to informational support — like looking up a technical manual or verifying an equipment specification. The actual execution, the part that delivers real value to the customer, still depends entirely on the professional’s hands-on skills. This creates an interesting dynamic where AI serves as a copilot, but the human remains the irreplaceable pilot.
Why hands-on skills hold up where AI can’t reach
To understand why hands-on skills have this natural resistance to automation, it helps to look at how AI models actually work in practice. Systems like ChatGPT, Claude, and other large language models are extremely good at processing text, identifying patterns in structured data, and generating responses based on verbal context. But they don’t have a body, they don’t have senses, and they can’t interact with the physical world on their own.
Industrial robots do exist and they handle repetitive tasks in controlled environments like assembly lines, but the cost of developing a robot capable of navigating a customer’s home, identifying where the plumbing problem is, and fixing it is still nowhere near commercially viable at scale. The variables are simply too many — every home has a different layout, every plumbing or electrical system was installed differently, and access conditions change from job to job.
Beyond the technology barrier, there is a massive economic barrier too. Even if it were possible to build robots that could replace electricians or HVAC technicians, the cost of manufacturing, maintaining, and operating those systems would be prohibitive for most markets. The skilled worker, with their tools and knowledge built up over years of experience, still represents a far more efficient and accessible solution. And that’s not even counting the fact that every electrical installation, every plumbing system, and every climate control unit has its own quirks — dealing with those variables requires an adaptability that goes way beyond what algorithms can deliver today.
There is also a human element that tends to get underestimated in this whole conversation, and that is customer trust. When someone calls a technician to fix a problem at their home or business, there is a relationship of accountability and communication that is baked into the service itself. The professional needs to explain what they found, offer options, negotiate timelines and costs, and often reassure the customer during a stressful situation. That kind of emotional intelligence applied to a practical service context is something AI simply cannot replace — at least not the way the skilled trades market operates today.
The white-collar paradox: who is actually more exposed to automation
One of the most interesting takeaways from Anthropic’s research is the stark contrast between public perception and what the data actually shows. For decades, conventional wisdom said that automation was a threat to manual laborers — to people operating machines in factories or doing heavy physical work. The narrative was always that the more formal education someone had, the safer their position in the job market would be.
Generative artificial intelligence turned that logic on its head. Large language models are language-processing tools, and that means they compete directly with professionals who work primarily with textual and digital information. Writers, financial analysts, lawyers who review contracts, programmers writing repetitive code, consultants building slide decks — all of these profiles are now in a much higher exposure zone than a refrigeration technician who needs to figure out why a compressor is overheating in a cold storage unit.
This doesn’t mean those office jobs are going to vanish overnight. What the data suggests is that they will be profoundly transformed, with many tasks being absorbed by AI tools, reducing the number of people needed to perform them. In the skilled trades, that transformation is happening in a much milder way, and in most cases, it functions as an upgrade to the tools available to the worker — not as a threat to their job.
Which professions are in the safest group according to the data
Based on data gathered by Anthropic and complementary studies from organizations like the McKinsey Global Institute and the World Economic Forum, it is possible to draw a clear profile of the professions that fall into the lowest exposure group when it comes to AI automation. At the top of that list are skilled trade workers:
- Electricians — they deal with variable circuits, location-specific codes, and unpredictable environments
- Plumbers — they rely on sensory diagnostics and work in confined spaces
- HVAC technicians — they operate complex systems with constantly changing environmental variables
- Welders — they perform precision work that demands fine manual coordination
- Heavy vehicle mechanics — they diagnose mechanical failures in multi-variable systems
- Elevator technicians — they combine electronics, mechanics, and working at heights
All of these occupations share common traits: they operate in physical, variable environments, require extensive certifications and hands-on training, and depend on diagnostic reasoning that cannot be handed off to a remote system.
Outside the world of traditional hands-on skills, other professions with a strong human component and physical presence also rank well in this scenario. Healthcare aides, physical therapists, mental health professionals, early childhood educators, and specialized chefs are all examples of fields where AI automation has limited reach. The common denominator is always the same — the more a job depends on physical presence, contextual judgment, and direct human interaction, the lower the risk of being replaced by technology in the coming years.
It is worth noting that this doesn’t mean these professions are completely shielded from the impact of artificial intelligence. What changes is the nature of that impact. Instead of replacement, what we see is an evolution of the tools available to these professionals. An electrician can use an AI-powered app to look up updated technical codes. An HVAC technician can lean on diagnostics assisted by smart sensors. AI comes in as an ally, boosting productivity and cutting down problem-solving time, without taking the professional out of the center of the operation 🛠️
How AI can be an ally for skilled trade professionals
If artificial intelligence isn’t going to replace plumbers and electricians, it can definitely make them more productive. And that is an important distinction worth paying attention to. There are already apps that use AI models to help technicians diagnose equipment failures based on symptoms described by the customer or picked up by sensors installed in the system. Thermal cameras integrated with smart software can map heat loss points in a building, making it easier for an HVAC technician to propose improvements.
Work order management platforms with built-in AI help independent professionals and small maintenance companies organize scheduling, estimate material costs, and even predict which types of service calls will be in higher demand during certain times of year. This streamlines the daily routine without requiring the professional to become a tech expert. The interface is designed to be simple, practical, and focused on solving real, everyday problems.
Another interesting development is the use of augmented reality combined with AI assistants to guide less experienced technicians through complex procedures. A professional just starting out can, for example, wear augmented reality glasses that overlay step-by-step instructions while they perform maintenance on a piece of equipment they have never worked on before. The AI provides the knowledge, but the person doing the work, making the calls, and adapting to the environment is still the human.
What this means for anyone thinking about their career
If you are at a point where you are choosing or switching careers, the data on automation and the skilled trades offers a pretty concrete perspective for making that decision. Investing in specialized technical training — whether in electrical work, plumbing, refrigeration, welding, or any other field that demands advanced hands-on skills — means betting on a market that combines high demand, a shortage of qualified workers, and low risk of being replaced by AI.
In the United States, for example, the Bureau of Labor Statistics reports a growing shortage of skilled trade professionals, with median salaries that already surpass many traditional college-degree careers. The baby boomer generation is retiring in droves, and there are not enough people entering these fields to fill the gap.
In Brazil, the picture is not all that different. Demand for qualified professionals in electrical installations, industrial maintenance, and climate control systems grows year after year, while the number of people entering these careers is not keeping pace. This creates a real window of opportunity for anyone who sees the value of hands-on skills in a market increasingly dominated by the digital work narrative. The combination of technical expertise with the smart use of AI tools as support is probably the most valuable professional profile someone can build for the next ten to fifteen years.
The message that comes through from Anthropic’s data is straightforward: artificial intelligence is transforming the job market, but that transformation is not happening evenly. It hits professions built around information processing and repetitive cognitive tasks much harder than those that depend on physical presence, hands-on diagnostics, and manual execution. The skilled trades are not immune to change, but they are in a much more comfortable position than most people realize. And understanding that, with data in hand, is the first step toward making smarter choices in a world that is changing fast 🔧
