Bench the benchmarks

In the social sciences as well as in engineering it is common practice to use benchmarks as indicators of performance. Thereby, several countries or regions within a country are compared with respect to quantitative indicator. Let’s take employment ratios. A higher employment ratio, which includes many persons working few hours in part-time work, is different from a slightly lower employment ratio, but hardly any part-time employees.
The same rationale holds true for benchmarks of AI systems or the newer versions of agentic AI that are under construction in many fields. The paper by Yuxuan Zhu et al. (2025) proposes the ABC (agentic behavior checklist) for agentic AI developers. The reporting of benchmarks by such models should include (1) transparency and validity, (2) Mitigation efforts of limitations and (3) result interpretation using statistical significance measures and interpretation guidelines.
The aim of this research is to establish a good practice in establishing benchmarks in the field of agentic AI. The sets of criteria to test for is large and the focus of how the agentic AI treats, for example, statistical outliers much above or below the average i.e. (> 2 standard deviations from the average) assuming a normal distribution, is one case of application only.
We welcome the efforts to bench the benchmarks in the field of AI as is common practice in other sciences as well.

Pervasive waste

From time to time waste from so-called highly developed countries is making headlines and then it is forgotten again. Huge amounts of plastic waste gets shipped for example from the USA to Malaysia in containers regularly (NYT 2025-7-1). The dumping of waste in other countries where it is cheaper to waste the waste is a cynical practice. Not only is the potential for reuse and a circular economy disregard, the little control that is exercised how the waste is treated afterwards is neglected. Some might just end up in our oceans later on or find its way in our food chains. The recent discovery of lots of nuclear waste at 5000 m depth in the sea in another extreme example of this practice to dump waste affecting all of is when profits have been accumulated inn the hands of a few enterprises and states. Such external effects as they are called in economic theory are part of the standard economic thinking. The challenge is to detect such behavior, persecute or better prevent it. This calls on countries who produce the waste to check for the contamination potential and treat their own waste. Fukushima has lots of barrels of nuclear waste waiting. The pervasive nature of this waste will make it last for thousands of years. “Beggar thy neighbor” with your waste is a major default of our current economic and social model. It remains an unresolved puzzle why mankind continues to work towards its own extinction. (Image: Le grisou, Constantin Meunier, MRBAB, Brussels). 

Contextual Vision

The attempt to define a sociology of vision has had a hard time to build on hard evidence that vision may depend on context or in a broader sense your visual heritage. A standard definition of context in vision highlights the areas around a focal point. The findings by Krupin et al. (2025) show through the comparison of persons from very different populations that our vision depends on our cultural background. It is the social background and upbringing in a specific cultural setting, which determines what we see in an image at first sight.  The so-called Coffer illusion test (Deregowski 2017) reveals what we see in an image spontaneously and maybe after some longer staring at the image or doing it repeatedly, we learn to see that there is more to see than our original impression. Depending on our cultural heritage we might focus unconsciously on rectangular or round shapes in a geometric image. This fundamental finding questions the view that there is only one universal kind of vision common to all humans. In fact, there is variance around what we see and thereby how we perceive an image. This research provides a justification to delve also into the field of a sociology of the visual. Because of the common term in informatics “WYSIWYG”, (what you see is what you get), we might  spend more efforts on research of how human vision is shaped over generations or according to social background. We know that in some images different people see different things. What appears as a splendid opportunity for some, is a very risky situation for others. Eyes are so closely wired to our brain that inscriptions of vision on the brain functioning are quite likely. The plasticity of this process over the life course remains a crucial topic to understand the process(es) of how a person’s social background shapes her/his vision.

Global warming

The annual update of the global warming indicators (source: Earth system science data 2025-6) gives more reasons to worry about the future of our climate. The objective to limit global warming to +1.5°C, established at the Paris climate agreement in 2015, is no longer achievable. This is the hard evidence based on the global network of scientific data collections and their projections. Fossil fuels are a major cause, deforestation and greenhouse gas emissions as well (Forster et al. 2025). As a consequence, human-induced warming of average surface temperatures and flows of heat into oceans continue. The first consequences, we witness in many parts of the world already. France is particularly affected (Le Monde 2025-6-20, p.7). What used to be called “natural disasters” is better described as long-term consequences of human-induced global changes like global warming. About time to take our CO2 footprint even more seriously. The “All electric society” can reduce reliance on fossil fuels considerably, as of now.

Fertility Fecundity

The scientific debate around changes in fertility has focused on social, economic and cultural factors to explain the drop in total fertility rates in OECD countries. The baby boom years of the 1950s and early 1960s had come to an end following the spread of new forms of birth control like contraception from the late 1960s onwards. The trend is very obvious and yet, the explanations of the trend might lack a more profound analysis of fecundity in addition to the socio-economic explanations. Shakkebaek et al. (2025) point for example to the little known effects of environmental (pollution) factors on the biological reproduction capability of humans (men and women).
Additionally, psycho-social factors like “the German Angst” fear about future developments in many social and economic fields might have direct effects, but also indirect effects on human biology. We know still very little about such feedback loops or feedback effects. We are more convinced, than we actually have hard evidence, that the BSP, SPB or PSB (B=Bio, S=Social, P=Psycho) spheres are interwoven, but an ambitious research agenda is called for to enlighten the issue. The big invisible elephant in the room might be environmental issues that enter into the fertility equations more than we have expected for many years. A nice working hypothesis for an ambitious and overdue research agenda.

Technology maturation

Technology has its own time of maturation. The time for wind power generation was quite lengthy compared to some innovation cycles in information technologies. Maybe, due to the fact that large wind turbines had only big enterprises or public monopolies as potential clients who were already heavily invested in even bigger nuclear energy projects caused the slow pace of development of this technology. In Germany just like in Denmark, Sweden and the USA the first larger research projects took off in the mid or late 1970s. The technology is explained by Erich Hau in a comprehensive way including the early days like the « Growian » test wind turbine in Germany (extract of image below). Despite the fact that major engineering difficulties were overcome already in the early 80s, the economic and legal challenges took much longer to resolve. Nowadays, we have an additional case of energy sovereignty to add to the cost-effectiveness of power generation from winds offshore and onshore. In the coming years we are likely to see more and more decentral power generation from wind on roof tops or small poles to complement solar energy during cloudy but windy weather or at night. Just like the sunshine we shall praise the windy days for their power generation potential.

Deus DEI

In Latin the declination of deus = god starts with the genitive form “dei”. In the world of universities and labor markets the abbreviation “DEI” stands for diversity, equity and inclusion. These 3 topics have ensured for years that a university campus looks a bit more like society at large, even if biased towards younger generations. The MAGA saga of the USA in 2025 starts to turn the clock back on hard to accomplish advances in DEI-hiring and promotion. This likely to reduce the number of international students on US campuses and will make campuses look WEIRD again. Moreover, lots of talented students will seek DEI campuses in other countries to avoid discriminatory hiring practices and arbitrary risk of visa denial.  Europe is likely to benefit from this kind of brain drain and skill shortages might be substantially reduced due to such unprecedented shifts in US policies. The cuts of billions of grants to renowned universities for political reasons will leave a yawning gap of talent in the US in a few years time, the latest. Early movers will move very quickly to anticipate the exodus of talent.  It is puzzling to watch how quickly years of DEI policies might be undone with thousands of people and projects at a loss. Scholars of American history are reminded of the McCarthy era of political prosecution or the dismal period for science under the Nazi-rule in Germany. Americans seem to get much worse and extreme policies than they thought they voted for. “Deus Dei” also means “god of god” and in the current American political context the judgement of a god by another god has maybe a historical meaning. In the table of the Estonian town hall in Tallinn the inscription alerts rulers to beware of being judged as well. Sounds like a timely reminder to overly ambitious politicians.

Journalists Scientists

In the theories of democracy journalism has entered the stage mainly in form of a controlling mechanism that is part of an extended checks and balances system of democracy. Independent journalists serve as effective multipliers and critical commentators which are an essential element of a well functioning democratic society. Social media’s ability to reach large audiences without adherence to independent journalists’ principles pose challenges to the way journalists are perceived by the public. The border between journalism and influencer marketing of products and opinions is continuously under pressure. Voices not heard,  underrepresented or those with few resources for large scale communication face difficulties to communicate their views in a market flooded with information and marketing. 

Another challenge to journalists is the fast evolution of scientific research and knowledge which at times derives from complicated theoretical and/or empirical methods. Hence, the relationship between journalistic reporting and scientific rigor and need for details in reporting are hard to reconcile. This necessitates a continuous dialogue between scientists and journalists to be aware of potentials and limitations to the co-evolution of the two disciplines. In order to avoid a self-referential subset in each discipline, the demonstration of successful cooperation between the two disciplines reveals the complementary role in many real-world instances. Investigative journalists often rely on additional scientific expertise and scientists who uncover manipulations of any sort in their work have to cooperate to inform society about the topics. As with global warming or PFAS risks scientists get drawn more and more into the checks and balances role within democracies. Therefore in addition to the separation of powers claimed in the political literature of the enlightenment (executive, parliamentary and judicial) we added already the 4th one of journalism and should include (5th) independent scientific research and reporting into the essential principles of well functioning democracies. The conference at the BNF in Paris on the 28th of March has raised these timely questions in the best tradition of the ongoing process of enlightenment.

Augmented Cognition

With everybody talking about artificial intelligence we tend to overlook recent advances in augmented cognition. Just 2 examples of studies presenting new findings published in 2025 suffice to make that point. Automated versus manual driving have different effects on our brains. Automation makes us sleepy and inattentive with potentially dangerous consequences. Using EEG measures of neuromarkers the drop in attention shows up earlier than through measures of keeping one’s lane while driving. For jobs or duties that involve extended hours of attention the wearing of a mobile EEG-device could avoid lots of accidents. Drugs, alcohol or medication abuse would also show up before it is too late, I suppose (Scanion et al. 2025).
The 2nd study involved a robotic assistance to piano playing which achieved surprising training effects for learners of the music instrument. Objective measurement of speed and finger coordination might even allow overall accuracy to improve significantly. The technology around us and the one in direct contact with us, our behavior and performances shall have an impact on all of us sooner or later. We have to pose the question of how far we want to go along with these evolutions. Augmented cognition will not only be of interest for civil purposes, but military applications or dual use is evidently another option.

Aufklärung Enlightenment

The period of the enlightenment is a historically defined time period of the late 17th and the whole 18th century. Although often associated only with Europe as its origin, the American contributions and influences have been far-reaching as well, just thinking of the “Declaration of Independence“. This historical period is, however, most interesting through the unrivalled contribution and pivotal point to the history of ideas. The exhibition and catalogue on the “Aufklärung – Enlightenment” by the German Historical Museum in Berlin exposes many of the historical editions of books that “made the enlightenment”. Philosophical writings and readings (Immanuel Kant: Kritik der …) of the period (especially in German) are tough even for native speakers of German, unlike reading of French (e.g. Jean-Jacques Rousseau) or English (e.g. John Locke) contributions to the enlightenment. Maybe this is already one of the reasons why in France the revolutionary ideas were put into practice through population-wide political movements.
The historical period of the enlightenment came to a close in the early 19th century (1806 another pivotal year). The English or French terms of enlightenment or lumière maybe be interpreted in a way that other historical periods like the medieval ages were periods of darkness, even later periods again did not live up to the idealist prerogatives of the evolution of humanity. The German term “Aufkärung” has its etymological roots more in the action associated with “to enlighten”. This hints at the continuing process of the ideas of the enlightenment in modern terms the process of individualisation, tolerance and autonomy.
There many valuable short interviews shown in the exhibition and Jürgen Habermas contributed a short closing statement to the catalogue of the exhibition on the unfinished project of the “Aufklärung” inviting us to continue also with the “Kritik der Aufklärung”. After all the enlightenment did not save us from repeated events like “Apokalyses“.
(Image: Émilie du Châtelet, below book on Newton Physics in DHM Enlightenment exhibition 2025).