However, in a study of science labs I conducted with three colleagues at the University of Manchester, we found that the introduction of automated processes that aim to simplify work — and free people's time — can also make that work more complex, creating new tasks that many workers might perceive as mundane.
We examined the work of researchers in a subject known as synthetic biology, or synbio, for short, in the study, which was published in Research Policy. The goal of synthetic biology is to give existing organisms new capabilities. It is engaged in the development of novel medications, the growth of meat in the lab, and new methods for making fertiliser.
Advanced robotic platforms are used in synbio research to repeatedly transport a lot of samples. In order to examine the outcomes of large-scale trials, they also employ machine learning.
These in turn produce a significant volume of digital data. Digital technology are utilised to modify conventional techniques and ways of functioning in a process known as "digitalization."
Scaling up science while freeing up researchers' time to focus on what they would deem more "valuable" work is one of the main goals of automating and digitalizing scientific procedures.
However, contrary to what one might anticipate, scientists in our study were not freed from tedious, laborious, or repetitive work. Instead, the activities required of researchers were multiplied and varied by the employment of robotic platforms. This is due to a number of factors.
One of these is the fact that more tests and hypotheses (the scientific word for a tested explanation for an observed event) were required. The possibilities are increased using automated approaches.
The number of hypotheses that could be evaluated increased, as did the possibilities for making small adjustments to the experimental design, according to experts. The amount of data that needed to be verified, standardised, and shared increased as a result.
Additionally, robots had to be "trained" to carry out studies that had traditionally been done by hand. For setting up, maintaining, and managing robots, humans also required to learn new skills. To make sure the scientific method was error-free, this was done.
The output of scientific research, such as peer-reviewed papers and funding, is frequently evaluated. Cleaning, troubleshooting, and supervising automated systems, however, take more effort than duties that are normally rewarded in research. Since managers are the ones who would be ignorant of monotonous labour since they don't spend as much time in the lab, these lower-valued duties may likewise be mainly unseen.
These teachings could also be applicable to other professional fields. An AI-powered chatbot called ChatGPT "learns" from information found online. The chatbot responds to queries from internet users with responses that seem well-written and persuasive.
According to Time magazine, people in Kenya were employed to filter harmful text sent by the bot so that ChatGPT wouldn't send back responses that were racial, sexist, or unpleasant in other ways.
The creation and upkeep of digital infrastructure require several, frequently hidden labour practises. A "digitalisation paradox" might be used to characterise this phenomena. It questions the notion that as elements of a person's workflow are automated, everyone involved in or impacted by digitalization becomes more productive or has more free time.
Organisational and governmental initiatives to automate and digitalize daily labour are mostly motivated by worries about a reduction in productivity. However, we shouldn't accept productivity increase promises at face value.
Instead, we should reconsider how we assess productivity by taking into account the various unseen jobs that people can complete in addition to the more obvious labour that is typically rewarded.
Additionally, we must think about how to organise and organise these procedures so that technology may enhance human talents more effectively.