A preliminary study by the American Federal Reserve rekindles the debate on the impact of artificial intelligence in tech and the crypto ecosystem. According to economists Leland D. Crane and Paul E. Soto, the rise of ChatGPT after November 2022 coincides with a sharp slowdown in developer employment in the United States. Over three years, this shortfall would represent approximately 500,000 jobs that would likely have existed without the rapid rise of large language models.
In brief
- A preliminary study from the Fed associates the rise of ChatGPT with a sharp slowdown in hiring developers in the United States.
- Employment growth in programming professions has reportedly fallen by around 50% since the end of 2022.
- Over three years, the gap would represent approximately 500,000 jobs that could have existed without the rise of major language models.
- Despite this decline, salaries have not fallen and AI-assisted programming could create new needs in the long term.
ChatGPT and a hiring shock measured by the Fed
Before the arrival of ChatGPT and the rise of generative artificial intelligence, professions related to programming were progressing at a steady pace. According to one study carried out at the Fed levelemployment in these functions was growing by around 5% per year, faster than the rest of the labor market.
Since then, the trend has changed. Job growth for programmers is estimated to have fallen by around 50%. In IT services and software development, it has even almost stagnated. The Fed sees this as a clear signal: generative AI does not only affect work tools, it also modifies hiring decisions.
However, the researchers isolated other factors. In 2022, tech suffered from rising rates, the end of the digital boom linked to the pandemic and the collapse of the cryptocurrency market. Despite this, their model estimates that programmer employment would still have fallen by around 3% per year, even after correcting for these effects.
A concentrated impact on code professions
L'analysis does not make ChatGPT the sole cause of the slowdown. Rather, it shows that generative AI creates a targeted shock on coding professions. Developers therefore remain among the most exposed profiles, because their tasks can partially be automated, notably coding, debugging and software architecture.
The researchers compare actual employment to a scenario where the share of programmers would have remained stable in each sector. Over three years, the gap reaches around 500,000 jobs, without representing a direct count of positions lost. Some of the workers concerned have probably joined nearby sectors.
The calendar reinforces this reading. The job deficit appears especially from mid-2024, around 18 months after the launch of ChatGPT. Companies have therefore undoubtedly tested these tools before reducing or slowing down their recruitment.
Junior developers seem more affected than senior profiles. Another Harvard study covering 62 million automatic data processing workers goes in the same direction. She observes a 9% to 10% drop in entry-level employment after the adoption of generative AI. In crypto, already weakened since 2022, this trend may increase the caution of technical teams.
What developers can expect in the coming months
The authors remain cautious. They are not talking about a lasting collapse. Programmer salaries did not fall, and job offers stabilized in 2024 before starting to rise slightly again.
Thus, ChatGPT could also create new needs. If AI-assisted code reduces certain costs, it can open up more projects, notably in cybersecurity, blockchain infrastructures, smart contract audits or crypto compliance tools.
However, the main issue concerns the talent pool. If entry-level positions become rarer, entry-level developers will have fewer opportunities to learn on the job. In the medium term, this could impact professional mobility and the training of future experts.
The Fed calls its work a first step. The study remains preliminary and has not yet been fully reviewed. However, it gives institutional weight to a trend already visible in several technology companies.
In conclusion, ChatGPT does not signal the end of the code profession. Rather, it accelerates a reorganization of work, where developers will have to combine technical skills, business understanding and mastery of AI tools. This development also goes beyond programming, as shown by the arrival of OpenAI tools intended for doctors, sometimes presented as more efficient than humans on certain specialized tasks. For crypto as for tech, the market could therefore look less for volume, and more for profiles capable of producing quickly, cleanly and with control.
Maximize your Tremplin.io experience with our 'Read to Earn' program! For every article you read, earn points and access exclusive rewards. Sign up now and start earning benefits.
