Logo Universität Bremen, back to home
Places and motives of acts of solidarity in transnational working relations

Collective action and AI: the next stage in accelerating digital transformation

Since recently ChatGPT seems to be on everyone's mind. A chatbot based on an artificial intelligence which is able to recognize more complex questions and to answer them accordingly at a level such that educational institutions are now in turmoil about how to deal with it. In the world of work, another fear is present: the potential of AI replacing workers (Lane/Saint-Martin 2021). Against this background, the technology of AI could pose certain challenges for employee representatives and require new or adapted strategies. In our research context, the more specific question is what impact this might have on collective action and solidarity.

For the time being, the technology only refers to and reproduces existing knowledge and the greatest achievement is to capture the question so well that the answer is appropriate, detailed and, above all, possible at an unprecedented speed. The AI is not yet so far advanced that it would be able to generate its own knowledge, but it is able to rearrange knowledge in such a way that e.g. ChatGPT can even be attributed to low-threshold creativity, if the chatbot e.g. takes on a role that is given to it and then continues the conversation in accordance with this role.

In the world of work according to Eurofund (2018) digitalization comprises three vectors of change in working processes: 1) the automation of work by replacing labour and tasks by machines; 2) the digitisation of processes by using sensors and rendering devices; 3) the coordination of economic transactions by platforms. AI technology could foreseeably accelerate developments within all three vectors enormously. While automation as a specter was still manageable for employee representatives in the past insofar as failures in automated production chains entailed too great a risk and it therefore tended to amount to semi-automation, AI technology could significantly reduce this risk with its ability to grasp complex interrelationships. Thus, automation can once again be understood as a threat to the world of work in the form of rationalization.

In industrial relations, especially in the platform economy, algorithms already show that digital technology does not have to be a neutral instrument, but that it can be used by a company's management to organize work in its interests, while at the same time reducing the transparency of organizational processes (Schreyer/Schrape 2018). Thus far, the issue of copyright of ChatGPT shows that there is a lack of transparency. Even though the chatbot largely only reproduces knowledge created by others, we as users do not get any source of information, for example.

In the transnational context of collective action, it seems obvious that AI, as a driver of technical translation being capable in replacing human translators (Bouguesmia 2020), could build bridges across language barriers. This can support cross-border working employee representatives in their exchange and communication, and fosters building trust which we consider fundamental for acts of solidarity.

If we ask ChatGPT directly: “What problems can a lack of transparency of AI bring within the working world?”, we get the following list of answers:

  1. Bias and discrimination in decision making.
  2. Lack of accountability and trust.
  3. Difficulty in detecting and mitigating errors.
  4. Ethical and legal challenges.
  5. Potential loss of jobs due to automation.
  6. Unintended consequences and fairness issues.

When asked how trade unions and other employee representatives should address this issue, ChatGPT provided the following responses:

  1. Advocate for transparency: Trade unions and similar actors should advocate for increased transparency in AI systems, to help address the problems outlined above.
  2. Represent the interests of workers: Trade unions should represent the interests of workers and ensure that their rights and interests are protected as AI becomes increasingly prevalent in the workplace.
  3. Promote upskilling and reskilling: As AI threatens to automate certain jobs, trade unions should advocate for programs to upskill and reskill workers, to help them transition to new roles.
  4. Address ethical and legal concerns: Trade unions should raise awareness about ethical and legal concerns related to AI, and work to ensure that these concerns are addressed in relevant policies and regulations.
  5. Collaborate with employers and policymakers: Trade unions should collaborate with employers and policymakers to ensure that the deployment of AI is done in a responsible and equitable manner.
  6. Monitor developments: Trade unions should stay informed about developments in AI and be prepared to respond to new challenges as they arise.

This response is not indicating a revolutionary new obligation for employee representatives but shows that ongoing developments in how to deal with digital transformation will increasingly shape trade union work. As for example the ongoing legal disputes around platform driven companies show (Daugareilh et al. 2019), AI just like algorithms must not be misunderstood as a neutral, because technical, tool in the context of required regulations. The social partnership is needed to accompany this, but also the transformational processes that are taking place with regard to workers resulting in (re)standardisation and dequalification (Butollo et al. 2018) remain a central task in the social dialogue. On the other hand, AI can create new jobs (Su et al. 2020), even if these will mainly be skilled jobs, which is why upskilling must remain a focus on the part of trade unions (Jaiswala et al. 2021)

At the current state of AI, human or social contexts, such as solidarity, do not play a role. However, the impact of AI as a technology in terms of its importance in labour relations must be viewed as an ongoing process. With this in mind and not least because AI is programmed and used by humans, it needs a regular evaluation by workers representatives. AI will be able to facilitate work processes, but at the same time it will always represent the risk that workers could be replaced. At this point in time, it is still the familiar patterns of known challenges that continue to dominate the relationship between employers and employees.

Acts of solidarity in transnational collective action needs bonds and bridges to forge alliances or even common identities (Nussbaum et al. 2022) that are able to demand rights and create political attention. AI in the world of work does not turn this on its head, but still requires regulation to prevent abuse. What could make AI special is its increased speed.

 

Literature:
Bouguesmia, Mohammed Tewfik (2020): Using AI in Translation, a Technological Leap, or a Translator’s Nightmare 2020 In: ALTRALANG Journal Volume: 02 Issue: 02 / December 2020. doi.org/10.52919/altralang.v2i02.77

Butollo, Florian/Engel, Thomas/Füchtenkötter, Manfred/Koepp, Robert/Ottaiano, Mario (2018): Wie stabil ist der digitale Taylorismus? Störungsbehebung, Prozessverbesserungen und Beschäftigungssystem bei einem Unternehmen des Online Versandhandels. AIS-Studien, 11(2).

Daugareilh, Isabelle/Degryse, Christophe/Pochet, Philippe (2019): The Platform Economy and Social Law: Key Issues in Comparative Perspective. ETUI Research Paper - Working Paper 2019.10, Available at SSRN: http://dx.doi.org/10.2139/ssrn.3432441

Eurofound (2018): Representativeness of the European social partner organisations: Commerce sector, Dublin.

Jaiswala, Akanksha/Aruna, C. Joe/Varma, Arup (2021): Rebooting employees: upskilling for artificial intelligence in multinational corporations. In: The InternatIonal Journal of human resource management 2022, Vol.  33, no.  6, 1179–1208; doi.org/10.1080/09585192.2021.1891114

Lane, Marguerita/Saint-Martin, Anne (2021): The impact of Artificial Intelligence on the labour market. What do we know so far? OECD Social, Employment and Migration Working Papers No. 256 doi.org/10.1787/7c895724-en

Nussbaum Bitran, Ilana/Dingeldey, Irene/Laudenbach, Franziska (2022).Theoretical conceptions of transnational solidarity in working relations. Bremen. www.iaw.uni-bremen.de/f/980b315b99.pdf

Schreyer, Jasmin; Schrape, Jan-Felix (2018): Algorithmische Arbeitskoordination in der plattformbasierten Gig Economy: Das Beispiel Foodora. In: Arbeits- und Industriesoziologische Studien 11 (2).        doi.org/10.21241/ssoar.64877

Su, Zhan/Togay, Guillaume/Côté, Anne-Marie (2020): Artificial intelligence: a destructive and yet creative force in the skilled labour market. In: Human Resource Development International 2021, VOL. 24, NO. 3, 341–352. doi.org/10.1080/13678868.2020.1818513

 

Updated by: Marcus Franke
RSS
Print page