The future of work: How will junior programmers be affected?

“Making predictions is difficult, especially about the future”. This quote, attributed to Nobel Physics Laureate Niels Bohr, perfectly reflects the caution with which statements about the impact of AI in the future should be considered. This is why it becomes relevant to rely on data from rigorous studies.

When confronted with the current debate on the impact of Generative AI (GenAI), we should be speaking more about jobs being affected by AI rather than replaced by it. In fact, according to the recent ILO report “Generative AI and Jobs: A Refined Global Index of Occupational Exposure”, 1 in 4 jobs worldwide is potentially exposed to GenAI, but tasks are more likely to be transformed than fully automated. 

It is true that the so-called “intelligence” of GenAI models has increased exponentially in recent years, to the point that there are Large Language Models (LLMs) that surpass the human level in performance tests based on doctoral-level questions. But one thing is the ability to solve specific problems thanks to the vast knowledge acquired during their training, and another is the real, effective and practical ability to fully replace a worker in the performance of all the assigned duties.

Furthermore, the real impact of AI will depend largely on the speed with which this technology is implemented globally. The lack of digital skills, the technological cost associated with the implementation of these constantly evolving technologies or the simple access to an adequate electricity grid or broadband Internet, are fundamental barriers that will necessarily delay the adoption of AI and, with it, its impact on the labour market. In this context, an accurate analysis of the effect that gaps in digital infrastructure could have in Latin America was carried out in 2024 in a joint working paper between the ILO and the World Bank.

Junior programmers in the spotlight

The discussion about the effects of AI on the labour market is an old one, although it has gained momentum after recent statements by Dario Amodei. The CEO of Anthropic (the enterprise behind models like Claude 4) considers that AI could eliminate up to half of the positions available for administrative jobs and raise unemployment by 10 to 20 per cent in the next 1 to 5 years. 

Traditionally, it was believed that automation would primarily replace manual and repetitive industrial jobs, leaving intellectual roles unaffected. However, the explosion of GenAI models following the launch of ChatGPT in November 2022 marked a significant paradigm shift: white-collar workers seem now more vulnerable than their blue-collar counterparts. 

As the aforementioned ILO report pointed out, while clerical occupations exhibit the highest level of exposure to AI, highly digitalized cognitive jobs (like web developers and software-related occupations) have experienced a higher increase in their mean scores since 2023. This trend can be attributed to the rapid advancements of LLMs, which were initially designed as sophisticated text generators, but are now capable to perform a broader range of tasks with the advent of multimodality. Multimodality extends LLMs capabilities far beyond text processing, enabling them to handle images, videos, audio, and code as both inputs and outputs.

It is paradoxical that the field of computer science, responsible for the great advances in AI, is also one of the sectors experiencing the fastest-growing exposure to its impact. This trend is particularly evident in entry-level positions, as the number of tools that simplify the tasks of a junior programmer (such as writing or debugging simple code) has multiplied since last year. This has sparked concerns among those who view these tools as a direct competitive threat, potentially leading to a decline in hiring young professionals starting a career with low value-added responsibilities. These concerns feel even more tangible if we consider the progress of agentic-driven technology, which enables the accomplishment of complex and lengthy tasks in a more autonomous way. 

Navigating the uncertainty: Advice for Junior Programmers

Firstly, while layoffs at tech companies are a reality, with news about “number Y of tech employees being dismissed Company X” every month, that does not necessarily imply the existence of a trend toward massive displacement of workers. On the contrary, as companies become increasingly digitized, a greater number of technical profiles with IT skills will be needed, even if their skills will need to be transformed to encompass new fields such as AI or big data. And this conclusion is one that can be easily drawn from the employment projections made by the U.S. Bureau of Labor Statistics. According to these projections, employment of software developers in the U.S. is expected to increase 17.9 per cent between 2023 and 2033, much faster than the average for all occupations (4.0 per cent), while employment of database administrators and database architects would also grow faster than average in the same period (8.2 and 10.8 respectively).

Secondly, provided that there will be room for more technical profiles with IT skills, junior programmers should strive to improve their competitiveness in order to offer more added value to the tasks they are typically assigned in entry-level positions. 

Lack of professional experience is a double-edged sword: while it can be an obstacle when tackling complex tasks and making important decisions, it also allows for faster and more organic adoption of new technologies, as there are no acquired habits that can slow down innovation. This advantage should be leveraged by junior programmers, using and exploiting the improvements that AI is bringing to the field of computing to improve their productivity and become valuable resources for any company that hires them. 

Thanks to AI, learning new programming languages or debugging one's own code is easier than ever, allowing junior programmers to gain professional experience (needed to become seniors) more quickly and effectively. In fact, and even if less programmers are needed, there would be a demand of Big Data Specialists, AI and Machine Learning Specialists and Software and Application Developers, high-related fields that, as we have seen, are expected to grow in the incoming years. The use of AI may facilitate the acquisition of new skills and the professional transition to different roles and profiles.

The main influence of technologies like GenAI on skills may stem from their ability to enhance human skills through human-machine collaboration, rather than fully replacing them, especially considering the ongoing significance of human-centred skills. 

If this is the case, there are no excuses for not using AI to learn, improve our productivity, and acquire new skills that increase the value of our professional profile.