10 occupations that AI automation cannot replace (Yet!)

Empathy role most secure against automation

A new study by Eskimoz evaluated the resistance of various occupations to automation driven by artificial intelligence (AI). The research, which assessed jobs across the UK, highlights those professions that remain most reliant on human interaction and are less likely to be overtaken by AI.

The study used two primary metrics to determine AI resistance: the percentage of public interaction required and the likelihood of automation.

The AI resistance score for each occupation was calculated based on these factors. Higher public interaction correlates with a greater dependence on human engagement, and a lower automation risk indicates a reduced likelihood of AI replacing the role. The findings were summarised by ranking jobs from most to least resistant to AI takeover.

Lawyers

The legal profession tops the list, with lawyers most resistant to AI automation. Lawyers require 100% human interaction, making their jobs immune to AI replacement. With a low 26% automation risk, the AI resistance score for lawyers stands at a perfect 100.

This is due to the profession’s reliance on complex human reasoning, interpretation of legal nuances, and personalised advice, which AI cannot replicate. Lawyers’ tasks, including client consultations, litigation strategy, and legal interpretation, require highly sophisticated judgment and human empathy.

Healthcare managers

The second most AI-resistant occupation is that of medical and health services managers, with an AI resistance score of 93. Although AI has a significant role in assisting with diagnostics and treatment suggestions, the human aspect of healthcare, especially in managerial roles, remains indispensable.

This occupation involves 89.8% interaction with the public, underscoring the critical importance of human compassion, communication, and decision-making in healthcare management. With an automation risk of only 29%, healthcare managers are unlikely to be replaced by machines in the foreseeable future.

Human resources managers

Human resource managers are in third place with an AI resistance score 87. These professionals interact with employees, job candidates, and other stakeholders in a highly human-centric role.

With 82.9% of the jobs requiring public interaction, HR managers handle complex interpersonal situations, such as conflict resolution, recruitment, and employee relations, all of which demand empathy and human judgment. Due to the need for nuanced decision-making in human interactions, the occupation faces a 26% risk of automation, similar to healthcare.

Operations and general managers

General and operations managers rank fourth, with an AI resistance score of 75. With an automation risk of 36%, this occupation remains relatively secure against AI disruption. These managers oversee various departments, teams, and projects, requiring them to engage directly with the public, staff, and clients. Over 80% of their work involves human interaction, emphasising the importance of leadership, decision-making, and strategic oversight—qualities AI struggles to replace.

Supervisors and administrative support

First-line supervisors of office and administrative support workers hold the fifth spot, with an AI resistance score of 64. This role requires 81.6% human interaction, reflecting the need for leadership, communication, and management skills in guiding teams and resolving issues.

While some tasks, such as scheduling and data management, can be automated, the job’s emphasis on people management and interpersonal skills makes it challenging to automate entirely. With a 50% risk of automation, this role faces some challenges from AI but is still far from being completely replaced.

Training and development specialists

Training and development specialists are sixth on the list, scoring 61 for AI resistance. With a relatively low 29% likelihood of being automated, these professionals develop educational content, training materials, and learning programs that require significant human interaction.

The role involves assessing trainees’ needs and providing tailored support, tasks that rely heavily on human engagement and insight.

Engineering and architectural managers

Seventh on the list are architectural and engineering managers, with an AI resistance score of 55.5. This profession has a relatively low automation risk of 25% but still requires considerable human interaction, with 47.1% of the role involving direct public communication.

The technical expertise and creativity required in this field make it resistant to AI disruption, but there is still some potential for automation in specific administrative tasks.

Compliance officers

Compliance officers are ranked eighth with an AI resistance score of 55.4. Although the job involves significant data analysis, risk assessment, and regulatory monitoring, the role still requires 72% human interaction.

Despite its high dependence on automation tools for tasks like document review, compliance officers’ expertise in interpreting complex regulations and liaising with stakeholders keeps them relatively secure—though with a 50% risk of being fully automated.

Industrial production managers

Industrial production managers hold the ninth spot, with an AI resistance score of 48.7. With a 37% risk of automation, this role requires 51.6% human interaction, as managers are responsible for overseeing production processes, solving problems, and maintaining communication with staff and clients.

While AI can assist in streamlining operations, the need for human oversight in managing complex production environments keeps this job relevant.

Graphic designers

Finally, graphic designers take the tenth spot, with an AI resistance score of 48.6. Despite AI’s growing capabilities in design, over 72% of the role still involves human interaction, especially in client communications and ensuring that designs align with branding objectives. While AI can assist in generating design concepts, the creative and subjective nature of the work ensures that human designers are still in demand, even though the automation risk is higher at 58%.

This comprehensive study highlights the varying degrees of AI resistance across different occupations, underscoring the importance of human interaction and judgment in the workforce. Although AI technology continues to evolve, jobs that rely heavily on human expertise, empathy, and decision-making remain the most secure against automation.

Image: The Eskimoz study’s findings were summarised by ranking jobs from most to least resistant to AI takeover. Credit: Pavel Danilyuk