Industry 4.0: Can AI ethics be embedded in the innovation lifecycle?
Lead Research Organisation:
University of Strathclyde
Department Name: Work, Employment and Organisation
Abstract
Can AI ethics frameworks be embedded in Industry 4.0?
Industry 4.0 (I4) refers to the digitization of manufacturing. Smart and autonomous systems that connect multiple machines make use of data to enhance manufacturing processes across the entire lifecycle of products-from concept and design through to use, maintenance and end-of-life. AI is critical to this. It enables computer systems to act autonomously to, for example, diagnose problems and to solve them. The role of humans in shifting. On the one hand, AI displaces the need for human input in decision-making. On the other hand, humans will work in ever closer connection with the AI-led data collecting tools, such as wristbands, data glasses, or other sensor related devices.
Given widespread ethical concern regarding AI and human machine interactions, the numbers of ethical frameworks in existence have proliferated in recent years. However, these tend to focus on abstract principles with little detail on how to practically apply them when designing AI or when implementing AI systems at different stages of the innovation lifecycle. Ethical frameworks targeted specifically at I4 tend to focus on employee health and safety, with little consideration of employee voice.
The project involves scoping work that will lay the foundation for future in-depth investigations into the impact of AI ethics frameworks on workplace cultures in I4. Our guiding questions for the scoping study are:
How do different stakeholders and drivers within I4 networks perceive ethical issues related to AI? To what extent do these perceptions reflect dominant cultures within the I4 innovation lifecycle which impact the implementation of responsible AI?
How are stakeholders' perceptions of AI ethics shaped by their own workplace culture and to what extent do they reflect underlying professional, occupational, or broader societal values and norms?
Our stakeholders are variously positioned in relation to the innovation lifecycle in I4. This includes teams tasked with taking such innovations through the lifecycle, such as those involved in design, testing and use. It also includes the teams developing the 'stack' of technologies, from hardware, compilers, libraries, services and application development, needed for AI innovations in I4. Aligning our focus with the full range of actors and actions taken in I4 will enable us to better capture the broad-spectrum workplace culture issues involved, including those arising from the interconnection between these actors.
The project will identify and form a Community of Interest (CoI) that brings together the range of stakeholders involved in I4. The CoI will center around our project partners, the National Manufacturing Institute Scotland, a government-funded hub that acts as a nexus of these stakeholders, and a data science company, providing the world's most widely used Python distribution, cloud services for industrial AI and data science and professional services for the development of AI/data science applications.
In our scoping study report, we will present a set of topics and themes to inform a future programme of research relating to workplace cultures and the challenges for embedding AI ethics in I4 innovation throughout the lifecycle. This report will emerge as part of a dialogue between our project partners, members of the Community of Interest, and the research team.
Industry 4.0 (I4) refers to the digitization of manufacturing. Smart and autonomous systems that connect multiple machines make use of data to enhance manufacturing processes across the entire lifecycle of products-from concept and design through to use, maintenance and end-of-life. AI is critical to this. It enables computer systems to act autonomously to, for example, diagnose problems and to solve them. The role of humans in shifting. On the one hand, AI displaces the need for human input in decision-making. On the other hand, humans will work in ever closer connection with the AI-led data collecting tools, such as wristbands, data glasses, or other sensor related devices.
Given widespread ethical concern regarding AI and human machine interactions, the numbers of ethical frameworks in existence have proliferated in recent years. However, these tend to focus on abstract principles with little detail on how to practically apply them when designing AI or when implementing AI systems at different stages of the innovation lifecycle. Ethical frameworks targeted specifically at I4 tend to focus on employee health and safety, with little consideration of employee voice.
The project involves scoping work that will lay the foundation for future in-depth investigations into the impact of AI ethics frameworks on workplace cultures in I4. Our guiding questions for the scoping study are:
How do different stakeholders and drivers within I4 networks perceive ethical issues related to AI? To what extent do these perceptions reflect dominant cultures within the I4 innovation lifecycle which impact the implementation of responsible AI?
How are stakeholders' perceptions of AI ethics shaped by their own workplace culture and to what extent do they reflect underlying professional, occupational, or broader societal values and norms?
Our stakeholders are variously positioned in relation to the innovation lifecycle in I4. This includes teams tasked with taking such innovations through the lifecycle, such as those involved in design, testing and use. It also includes the teams developing the 'stack' of technologies, from hardware, compilers, libraries, services and application development, needed for AI innovations in I4. Aligning our focus with the full range of actors and actions taken in I4 will enable us to better capture the broad-spectrum workplace culture issues involved, including those arising from the interconnection between these actors.
The project will identify and form a Community of Interest (CoI) that brings together the range of stakeholders involved in I4. The CoI will center around our project partners, the National Manufacturing Institute Scotland, a government-funded hub that acts as a nexus of these stakeholders, and a data science company, providing the world's most widely used Python distribution, cloud services for industrial AI and data science and professional services for the development of AI/data science applications.
In our scoping study report, we will present a set of topics and themes to inform a future programme of research relating to workplace cultures and the challenges for embedding AI ethics in I4 innovation throughout the lifecycle. This report will emerge as part of a dialogue between our project partners, members of the Community of Interest, and the research team.
| Description | What were the most significant achievements from the award? One of the key findings of this short term explorative research project was the realisation that AI might be talked about a lot, but manufacturing clearly is not on the forefront yet but has a way to go to implement and meet the criteria for what makes the Industry 4 revolution. Having said that, our research shows that the manufacturing communities clearly wish to engage with debates that are related to but not limited to AI and ethics. The broader context of ethics and manufacturing with regards to the percepetion of the 'human in the loop' are ongoing, and part of workplace cultures. Not the least the move towards using the notion of Industry 5 to underline the holistic approach incuding human centric, sustainable, and ethical apoproaches, is making an impact on communities of praxis. Another key finding hence is that the appetite for debate is there, but the integration into the process often remains no more than a lip service. In manufacturing, the innovation as process is still in the main focus, and human safety, health, or risks assessments are mostly designed around the idea on how a machine could improve to support the human. Rarely is the Operator 4/5 modelled in detail, or even seen as an employee with a voice, but more an assemblage of human factors. Engineering, Computer Sciences, and related disciplines are aware of this tension, and the project could facilitate some of these to become accessible for discussion. In this respect, one of the key aims for this preoject - creating a community of interest - is happening while at the same time leading this through our project did not match the timelines for industry partners and stakeholders. We had to realise that given the multiple layers of events and debates on AI ethics increasingly taking place makes industry question the 'added value' of a community of interest. These preliminary findings will be taken forward by the PI and the research team. We will apply for further funding, but more importantly continue to work on collaborations started already. Mainly the testing of using an AI ethics framework in real world smart manufacturing as planned with our industry collaborator is on the agenda still and will b erealisedin 2023. |
| Exploitation Route | Academic routes: We will apply for further funding to engage more with the 'gaze of the experts' in AI/ML and manufacturing, specifically with regards to design. There is evidence that ideas around inclusive work and a more holistic perspective of employees is needed, but so far a lack of concepts and models is hindering the operationalisation of innovative concepts. To understand the vision and versions of the human in manufacturing will deem to deliver solid research agendas and outcomes. Non-academic route: Clearly the focus on AI and workplace cultures in production and manufacturing is relevant for futures at work. Recents debates on warehouse work, decent work, or fair work can be supported with our first explorative approaches. |
| Sectors | Education,Government, Democracy and Justice,Manufacturing, including Industrial Biotechology |
| Description | Given the short term and small scale of the project, the impact narrative by necessity is rather difficult to assess with regards to measureables. Some impact in nucleus status might be named: a) Together with the interdisciplinary team, we managed to increase the importance to relate on AI ethics beyond simple 'lip service'. We could flag the importance of organisational structures, of hierarchies, and business needs to industry; with our engagemnet in education, the next generation of people who will be working in the field could be reached. The invite to collaborate from within STEM disciplines for us is an indicator for this impact. b) Our interdisciplinary approach allowed us start with the nucleation of a new research area that focusses on the 'creation of the human' not by machines, but by the makers, coders, engineers, and designers. |
| First Year Of Impact | 2022 |
| Sector | Education,Manufacturing, including Industrial Biotechology |
| Impact Types | Societal |
| Description | AI and workplace cultures. Workshop at Scottish AI summit 28-29th March 2023 |
| Form Of Engagement Activity | Participation in an activity, workshop or similar |
| Part Of Official Scheme? | No |
| Geographic Reach | National |
| Primary Audience | Policymakers/politicians |
| Results and Impact | At the time of submission the workshop was to happen but I decided it to be important to know that the workshop was accepted for the summit as a metter of esteem (the organisers mentioned only a third of the workshops submitted had been invited). |
| Year(s) Of Engagement Activity | 2023 |
| URL | https://www.scottishaisummit.com/ai-workplace-cultures |
| Description | Engage Strathclyde Event 'Industry 4.0: Can AI ethics be embedded in the Innovation lifecycle' |
| Form Of Engagement Activity | Participation in an activity, workshop or similar |
| Part Of Official Scheme? | No |
| Geographic Reach | National |
| Primary Audience | Industry/Business |
| Results and Impact | The event focused on manufacturing to understand the potential for AI ethics frameworks in the I4 innovation cycle, and the outcomes for human labour. Our event was aligned with newly developed strategies for ethics in AI, including the work by Innovate UK (via Catapult High Value Manufacturing), the Scottish AI Alliance, and the national Manufacturing Institute Scotland, NMIS. We engaged with the now leading industry framework, Aletheia, an ethics assessment toolkit developed by Rolls Royce with open access for the public. The aim of the event was to support building a Community of Interest in AI ethics in manufacturing. The attendance from industry leaders allowed the planning for a collaborative project with NMIS to test the ALETHIA framework. |
| Year(s) Of Engagement Activity | 2022 |
| URL | https://www.engage.strath.ac.uk/event/944 |
| Description | Ethics in AI Workshop. AI summit Fringe event at London Tech Week 2022 |
| Form Of Engagement Activity | A talk or presentation |
| Part Of Official Scheme? | No |
| Geographic Reach | International |
| Primary Audience | Industry/Business |
| Results and Impact | The workshop took place during teh London Tech Week, 16th June 2022. The PI for this project, Dr Briken, and Dr Michael Fisher had been invited to input on the topic to facilitate a debate. Interest was high specifically from policymakers on the issues of ethics in industry in general. |
| Year(s) Of Engagement Activity | 2022 |
| URL | https://london.theaisummit.com/speakers/kendra-briken |
| Description | Participation as guest in the series 'Free Thinking-New reserach' on Artificial Intelligence (for BBC R3, recorded 2nd March 2023) |
| Form Of Engagement Activity | A broadcast e.g. TV/radio/film/podcast (other than news/press) |
| Part Of Official Scheme? | No |
| Geographic Reach | National |
| Primary Audience | Public/other audiences |
| Results and Impact | With Kerry McInerney and Eleanor Drage, Research Associates at the University of Cambridge's Leverhulme Centre for the Future of Intelligence (Kerry is also a 2023 AHRC/BBC Radio 3 New Generation Thinker) the PI Dr Kendra Briken discussed AI, and kendra focussed on industry related issues. |
| Year(s) Of Engagement Activity | 2023 |