The AI skills gap is here, says AI company, and power users are pulling ahead

The debate over the impact of artificial intelligence on the labor market usually oscillates around apocalyptic visions of mass unemployment. Meanwhile, the latest data from Anthropic, one of the leaders in the generative AI sector, sheds entirely new light on this dynamic. Instead of a rapid wave of layoffs, we are observing the formation of a dangerous skills gap that is beginning to divide workers into those who can harness models for work and everyone else who is left behind.

Peter McCrory, Head of Economics at Anthropic, presented findings from the latest report on the economic impact of technology during the Axios AI Summit in Washington. Although the labor market remains "still healthy" in his assessment, processes are occurring beneath the surface of statistics that could permanently change the employment structure. The key conclusion is not a lack of work, but the progressive marginalization of individuals who have not adapted to the new tools.

The privilege of proficiency in the era of language models

The phenomenon that Anthropic calls the "AI skills gap" is becoming a new determinant of professional status. Data suggests that so-called power users — individuals proficient in using advanced models such as Claude 3.5 Sonnet — are gaining a disproportionately large advantage over their peers. This efficiency does not yet translate into job cuts on a macro scale, but it makes the barrier to entry for higher positions almost insurmountable for those without these competencies.

Analyzing this trend, it can be noted that AI currently acts as a productivity catalyst for already experienced workers. Instead of replacing humans, this technology empowers those who possess solid substantive foundations and can critically evaluate machine-generated results. This leads to a situation where one "AI-armed" specialist can perform work that previously required an entire team, which in the long run must affect the demand for new personnel.

• Growth of inequality: Experienced users gain an advantage in the pace of task execution and the quality of results.
• Employment stability: At the current stage, the report does not confirm mass layoffs directly caused by automation.
• Competency barriers: The lack of skills to operate AI tools is becoming a new type of professional exclusion.

Young workers in the crosshairs of change

The most concerning aspect of the Anthropic report is the technology's impact on those just entering the labor market. Peter McCrory points out that it is the youngest generation of workers who may feel the effects of the transformation most acutely. Traditionally, junior positions served to learn the craft by performing simpler, repetitive tasks. Today, these tasks are the first to be taken over by sztuczna inteligencja.

If juniors are not given the chance to learn the basics because the "dirty work" is performed by an AI model, the industry may face a massive challenge regarding competency continuity. Tech and creative companies must revise their mentoring systems. In a world where Claude or other Large Language Models (LLM) write code, summarize documents, and analyze data in seconds, the role of an intern must be redefined to avoid a generational gap.

"Currently, we do not see evidence of widespread displacement of workers from the labor market, but early signals indicate an uneven distribution of the benefits flowing from AI." — Peter McCrory, Head of Economics, Anthropic.

Efficiency that masks upcoming changes

It is worth taking a closer look at why unemployment statistics have not yet budged drastically. Many organizations are currently in an experimental phase. Companies are implementing solutions from Anthropic or OpenAI to increase margins, and not necessarily to immediately reduce personnel costs. However, this "healthy" phase of the labor market mentioned by McCrory may only be the calm before the storm, in which the optimization of organizational structures will occur.

The key term here becomes "augmentation" instead of "substitution." A worker who can integrate AI into their workflow becomes invaluable to the employer because their unit labor cost relative to the generated value drops drastically. It is this group of power users that dictates the pace of change today in sectors such as software engineering, digital marketing, or financial analysis. The rest of the employees, operating on traditional methods, become relatively more expensive and less competitive.

This phenomenon can be compared to the introduction of spreadsheets in the 1980s. Accountants did not disappear, but those who refused to learn how to use a computer were quickly displaced by a new generation of specialists. The difference is that adaptation to AI is happening orders of magnitude faster, and models like Claude 3.5 evolve in monthly cycles rather than decades.

A new paradigm of professional education

The conclusions drawn from Anthropic's research force a change in thinking about education and professional development. Since AI does not take away work but changes its nature, "AI literacy" — the ability to understand the capabilities and limitations of probabilistic models — becomes key capital. It is no longer just about the ability to write prompts, but about the strategic management of processes in which AI is one of the links.

The advantage of power users does not stem from access to better tools — these are usually widely available — but from a deep understanding of how to verify model hallucinations and how to connect different systems into coherent workflows. Companies that want to survive this transformation without drastic personnel shocks will have to invest in internal upskilling programs instead of hoping that the market will provide ready-made experts on its own.

In the near future, we will witness the deepening of this stratification. The labor market will not be destroyed by algorithms, but it will be brutally filtered through the prism of efficiency. The real threat is not a robot that will take your desk, but the colleague at the next desk who learned to collaborate with that robot twice as fast as you. Professional transformation is ceasing to be an option and is becoming a necessary condition for maintaining one's current economic position.