In the first few years of wider adoption of AI in an economy, there is the expectation that this might lead to substantial productivity gains for enterprises which use it as well as for employees who are early adopters of the relatively new technology. The study by the Stanford Digital Economy Lab by Chen, Chandar and Brynjolfsson (2025) showed that so far there are no significant earnings effects for employees. Based on millions of recent payroll data from US companies productivity gains have not trickled through to the paycheck in terms of monthly salaries. Participation of staff in a company’s overall turnover or profit might change this as time evolves. For civil servants the adoption of AI might mean increases in cases dealt with as some tasks can be executes faster than before with the use of AI.
The evidence points to employment effects of AI rather than earnings effects so far. A hypothesis is yet unresolved: senior employees using AI might employ fewer junior workers at entry positions, if these “hallucinating” young professionals can be replaced by hallucinating AI. In science the hallucination has sometimes lead to disruptive new approaches and findings. It is a tough choice to pick the young entrants with high productivity potential and eventually high remuneration for this in terms of labor earnings.
AI employment effects
The first robust empirical evidence about employment effects of AI in the USA has been published by the Stanford Digital Economy Lab by Chen, Chandar and Brynjolfsson (2025). A previous paper by Wang and Wang (2025) highlighted the comparative advantage of persons who use AI in their work compared to others and the authors coined the term “learning by using technology”. The prediction of the model was that there might be job losses of more than 20% in the long run and half of this already in the first 5 years of the introduction of the technology. The Stanford economists have estimated with real world data these effects in the USA and find quite surprisingly that the negative employment effects of AI have the strongest impact on young labor market entrants with few years of labor market experience. Middle-aged and more senior employees seem to benefit from “tacit knowledge” about the work, which is more difficult to replace with AI, at least for the time being of the early days of AI. This evidence is based on recent payroll data from the largest payroll processing firm “ADP” in the USA which has firms overrepresented from the manufacturing and services industries as reported in another paper (Firm size maybe another source of bias). However, the effect that youth 22-25 years of age suffered the most calls into question the common belief that older workers are more likely to suffer the consequences as during in the rise of the digital economy around the year 2000. (AI Image created with Canva)
Vacation or Workation
Vacation or Workation, that is the question. Remote work and outsourcing of work have created the opportunity for more people to work from distant places. Logging into the firm’s or the administration’s secured intranet enabled to work from where you prefer to work rather than on the premises of the firm. After the technological shift to enable secure remote work, the acceptability of remote work is a societal issue. Whereas bosses worked or had to work while on travel for many decades, the same has become feasible due to reduced costs for many other employees as well. The Covid-19 pandemic has forced another shift in the need to accept and advance remote work as well. Taken together the option of a “workation” instead of a vacation became equally more feasible. The borderline between remote work, “workation” and vacation has become more and more blurred. It remains to be seen, whether these options enhance an outsourcing trend by firms or whether more “back-sourcing” or “in-sourcing” will be the consequence.
Employers with their associations and employees with their trade union representatives have a need to include such topics in their bargaining agenda in addition to pay, health and safety, as well as working time adaptations.
(Image created with Canva 2025-9)
Work time reduction
One of the major elements of social progress in the 20th century consisted in the reduction of work time. Reductions from 48 hours per week in the first half of the century were largely reduced to 40 hours per week or less in some industries with strong trade union representation. State regulations also pushed in this direction with positive implications for physical and mental health as well as wellbeing. Advances in longevity of employees may be attributable to this social progress agenda of the 20th century. In the 21st century we witness a new thrust of enterprises and employees striving to implement a 4 day work schedule by at the same time organizing a further reduction of work hours. The scientific evidence which is based on pre- and post trial assessments of workers satisfaction shows rather positive results (Fan, Schor, Kelly, Gu 2025). More studies are due to accompany this potential of further health and wellbeing effects of reduced work time and the reorganization of work time in enterprises.
Electrifying Jobs
The transition to the “all electric society” necessitates to prepare the labor force for the upcoming challenge. The knowledge about electricity and electric appliances constitutes the basics of the knowledge base of the future. This goes far beyond the basics of physics and electrical engineering. There many processes like sharing of electric infrastructure in households, cities, in and beyond countries that have to be delt with. Investment calculations and legal issues to address the different risks involved are another area to cover in the process to prepare society for the “all electric society”.
However, the skills of professions with more direct links to the fossil fuel based technologies have a role in the phasing out of the heavy reliance on fossil fuels. Reverse engineering of such engines and heating will need people still knowledgeable of the past, when younger generations set their focus primarily on professions with links to electricity. Even using a solar powered heat pump in a home requires pipes to the existing network of radiators, for example. This will most likely be a gradual shift of the job structure and occupational requirements over at least a decade, but the shift has started already. Some might argue we need a well functioning “transitional labour market”, labour market policies and social security system for this to happen smoothly.
Work and Time
The link between work and time has been evolving as a major element of social progress over centuries. In what younger generations seem to take for granted, <40 hours week, paid time off-work. 4-5 days per week to name only the major ones, has been the result of huge struggles and hustles driven by employees and their trade unions to achieve such milestones. The current debate to get employees back into offices and/or to work longer hours again, is also in the end a debate about work and time. The advantage of working from home consists among other factors in saving a lot of valuable time and stress from commute to work. This unaccounted time to go to work has become a major health hazard even in an apparently comfortable company subsidized car or any other means of transportation. The traffic jams of so-called rush hours, when in fact everybody is slowed down, are a serious health hazard usually ignored by employers. However, these hours are a major part of the work-life-time balance of employees. In many negotiations and collective bargaining about working time this is the big elephant in the room rarely addressed.
AI 2nd round effects
The most popular topic currently is AI.
Most writers, assisted by some form of AI, will deal with the 1st round effects of AI. These consist in the immediate consequence of the use of AI in office work, medical and military applications, music and all producing or creative industries. As an economist you take the input – output matrix of the economy (OECD countries) and take AI as an additional dimension of this I/O matrix, for example. The result is an AI-augmented model of the economy. This 3-dimensional cubic view of the economy asks to reflect on the potential short-term and medium-term impact of AI.
Let’s take the example of translation and editing services. AI will in the short-term or the 1st round effects make it easier to offer mechanical translations with fast turnaround. Most likely, this will lead to less translators needed for routine translations of longer texts, which would otherwise be a very costly endeavour. The 2nd round effects, however, will make the expert knowledge of translators of texts, where every word counts, more necessary in order to provide the best version of a translation targeted on specific audiences.
In the legal domain, for example, the precision of words is primordial and errors can be very costly. Hence, the 2nd round effects of AI in this field will increase the demand for high quality translation services more than before the use of AI. The important shift consists in these 2nd round effects of AI, which give a push to multilingual societies as just one medium-term outcome.
Please use AI to read (listen) to this paragraph in your native language or even dialect using your favourite AI-tool.
AI Workday
Our workdays have seen considerable changes throughout the last few days. The home office boom has allowed employees to work for extended hours from home. The there is an abundant literature on the effects of home office work on well-being or the work-life balance. Productivity gains could be reaped by employers and a better work-life balance was a lasting advantage for employees.
The increased use of AI specific to some occupations has introduced a new form of added productivity for some occupations or professions, AI as complementarity, whereas other occupations suffered a higher risk of being substituted by AI applications.
Based on time diary data, the study by Wei Jiang et al. (2025) reports that users of AI have longer work time and reduced leisure time. Competitive labor markets increase the pressure to put in even higher hours of work. Nerds, just like workaholics, are likely to be drawn into excessive hours of work with increased health risks. Enterprises and consumers appear to be gaining more than the employees, who are at a higher risk of loosing out on their work-life balance over time.
Retirement Plans
There are debates about the best organization of retirement. The major fault lines lie between public pension systems and systems that are built based on mainly private provision. Retirement plans in either system are subject to constraints. The recent stock market turbulence has increased the amount of uncertainty people face who invested in 401K plans in the US. Some had to take an unannounced hit to their retirement savings due to the loss after Trump’s back and forth policies on tariffs (OECD Pension Outlook). High volatility of stock market prices creates an additional constraint that you are less inclined to retire when your retirement investments have overall a reduced value. You are a bit at the mercy of capital markets even in your retirement decision, irrespective of the difficulty to predict what your retirement funds will yield as returns. Quite an important lesson to keep in mind when comparing retirement systems in OECD countries. It has been all too easy to blame public pension systems for maybe lower short term interests on pension savings. Being subject to an American president concerning your retirement plans is probably not what many countries would like to have. Trump’s choices on tariffs may have consequences we did not expect to affect us so directly.
Composing Assisted
Before the existence of digital composition tools composers were assisted by “Kopisten”. These persons rewrote the original draft of a composition into a “proper” version of the original document. Musicology has a tough time to deal with deviations from the original. It needs to be clarified which is the final and authorized version. In some instances this is far from evident. Just as an example Robert Schumann made ample use of the assistance of Kopist Otto Hermann Klausnitz (cf Nr 6), sometimes for the preparation of the composition, the finalized versions or the explicit drafting of different voices. Klausnitz himself was a flautist in Leipzig (Gewandhausorchester) and a conductor in Duesseldorf. Overall the debate is still going on, whether the composer’s draft or in many instances the Kopist’s version of the composition (authorized or not) prevails. In the age of AI, which is highly influential in modern music, such questions will most likely be intrinsic to the process of composition as well. AI is influential in evening out rough edges. Anette Mueller (2010) has done a great job to make this work of “Kopisten” much more transparent and her concluding chapter is programmatically entitled “Komponist und Kopist- Aspekte einer produktiven Kooperation”. (Image Mueller, A 2010 p. 340).

AI Ghost Writer
Yes, with AI we have entered a new phase of the impact of IT. Beyond the general applications like ChatGPT there is a rapidly expanding market of AI applications with more specialized functions or capabilities. In the realm of scientific writing AI-Writer is an interesting example of the AI assisted production of scientific tests. After the specification of the topic you will receive several options to specify the content of the short paper you want to produce with AI-Writer. You may choose the headline, keywords, subtopics and the logical order of these subtopics depending on your audience. Alternatively, you leave all those decisions to the application and restrain yourself to fix the amount of words you would like the paper to have.
AI-Writer is a powerful ghost writer for much work even of advanced scientists. The quality of the paper needs to be checked by yourself, but the explicit list of references, from which AI-Writer derives its restatements of the content, is just next to it. Your ghost writer AI is likely to replace a number of persons that were previously involved to just produce literature reviews or large parts of textbooks sold to millions of students.
A much lesser known feature of such tools is the way it makes plagiarism much more transparent for the scientific communities and the public at large. These programs demonstrate the techniques of combining knowledge and the citation imperatives in a transparent, almost pedagogical way. This latter function will speed up scientific work like dissertation drafting, since the reading up and documentation of previous literature in a field is a time consuming early stage of academic degrees.
Email composition, rewording, plot generator or social media posts are additional nice-to-have features of the new AI-assistants. A lot of work that has been outsourced, for example, to lawyers, consultants or other technical professions, might equally be challenged. Ghost writers have been around for centuries. With AI for everybody, they will also be involved everywhere.
(Image screen shot of working with AI-Writer 2024-6)
Work Subordination
One of the defining principles besides the job description, working time, working place, remuneration of a work contract is the subordination to a superior. The employment contract entails some more or less strictly executed form of direction, but a right of direction nevertheless. This element of subordination has become a major issue in the definition of whether you are effectively an employee or a self-employed person. The digital revolution had enlarged the kinds of subordination. Platforms and algorithms, that distribute work among several employees (but named platform workers) disguised the subordination to a superior level of management to the platform and its seemingly anonymous algorithm. Many young riders were saying, I don’t like bosses, but I am willing to accept a “technical” platform that distributes work tasks only seemingly in a non-discriminatory way. Due to a failure of labour legislation to regulate early enough a thriving model of fake self-employment developed throughout Europe and beyond. Labour Courts have contributed a lot to correct the disguised subordination. Even Uber is advertising that they only operate as a broker of tasks, but have no subordinated employees. The related issue of subordination remains largely the same. Subordination to an algorithm of the distribution of tasks is the end result.
Many start-up enterprises use Kanban boards to facilitate project and team management. Shifting tasks between employees, introducing new tasks and self-selection of tasks are potentially subordination-free allocation of tasks (example software). Flat hierarchies seem more manageable through the use of such tools. The number of tools that integrate other office functions is impressive. When testing such tools, that become more popular also in the distribution of household tasks, beware of the data-hungry nature of such tools. For example, https://trello.com/ warns correctly in the small print that for its full functionality it would need to have permission to use “your” camera, microphone, contacts, photo library, calendar etc.
This demonstrates that subordination, nowadays, is complemented by the algorithmic use of a lot of privacy, we would never have agreed to a boss in person should even ask for. The new and old subordinators have powerful tools at hand, the subordinated will have to get their act together and limit the amount of subordination they are willing to accept.
Again, this is a generational topic. The low threshold to accept technical subordination in younger generations, your autonomous level-5 car is breaking earlier than you even perceive a risk, is confronted with the higher threshold to accept personal subordination for youth. Interestingly, for older generations the obverse is true. All in all, we have ample reason to rethink and re-define also in legal terms the manifold, disguised and new forms of subordination related to work.