The OECD has released its first Recommendation on AI as part of its annual Ministerial Council Meeting, recognising that a “fair transition” is needed for workers affected by AI deployment with social dialogue as means towards achieving it. The OECD AI Principles recognise trade unions as a relevant stakeholder and call for a responsible use of AI at work under the premise of inclusive growth and sustainable development.
The OECD Recommendation is the first set of international standards on AI developed in a multi-stakeholder setting – involving the TUAC in cooperation with UNI Global Union. They are signed by all OECD members – as well as Argentina, Brazil, Colombia and Costa Rica.
The TUAC welcomes that the Principles include obligations to all AI actors and stakeholders “for the responsible stewardship of trustworthy AI”, and provides a roadmap for public policy action.
UNI Global Union General Secretary Christy Hoffman said, “The application of A.l. should not be self-regulated and left to the big tech companies. The principles announced today create a much needed regulatory framework for national governments and underline the need for action.”
UNI’s Director of Platform and Agency Workers, Digitalization and Trade, Christina Colclough, added, “I am proud to have been part of this work with TUAC. We must welcome this ground-breaking move by the OECD to set the bar for international standards on AI. It’s a significant step forward. We stress the necessity of worker wellbeing, of social dialogue, of prioritising people and planet and not just profit. We must now use this to hold our governments accountable and committed to putting action into these principals.”
The principles are a step towards seriously addressing a reality of a few actors capturing most of the AI market. Policy makers and stakeholders, including unions, need strong frameworks to navigate this complex and changing space. Importantly, now the OECD itself and its members have to showcase their commitment through concrete actions on valuable outcomes such as:
- Following a human-centered approach;
- Linking AI design, deployment and use to human rights, the SDGs and existing instruments that safeguard responsible business conduct and decent work;
- and – very importantly – moving towards a fair transition for workers affected by AI and using social dialogue as a means to achieve it.
Labour market related principles in the OECD Recommendation on AI:
Governments should take steps, including through social dialogue, to ensure a fair transition for workers as AI is deployed, such as through training programmes along the working life, support for those affected by displacement, and access to new opportunities in the labour market.
Governments should also work closely with stakeholders to promote the responsible use of AI at work, to enhance the safety of workers and the quality of jobs, to foster entrepreneurship and productivity, and aim to ensure that the benefits from AI are broadly and fairly shared.
It is important that now with the Recommendation adopted to continue developing a better understanding of the effects of AI on our economies through an inclusive growth lens with the help of the AI Observatory – on which trade unions should be represented.
Going from there policies need to be concretely developed fostering built-in security and ethics, accountability and transparency in AI systems that lead to fair, non-discriminatory outcomes for users, including the workforce. For now, algorithmically-powered systems are often opaque, create bias and security risks for workers.
AI is a major trade union concern from different angles that need more attention:
- Algorithmic transparency for example on price setting that affects wages on online platforms;
- Liability of machine behaviour where it affects workers’ physical and mental health and towards safeguarding their autonomy and self-determination;
- Data protection of workers data and standards on how performance is monitored through AI; systems (for example through AI augmented robotics, smart gloves or sensors)
- AI induced bias in human ressources (HR).
To make AI improve working conditions, it is a key requirement to let workers and their representatives participate and negotiate the goals of the use of technology from the very beginning. The bottom line is that workers need to be informed and understand any workplace and organisational decisions made on the basis of AI affecting their tasks, income, health and safety. Trade unions, works councils, safety and health organisations and training bodies need to be involved in future policy discussions and decisions – also by strengthening bargaining and co-determination rights. Social partners and governments need to make sure that productivity gains of AI are adequately measured and shared – and not ring-fenced in a for now concentrated market.
Note to editors: UNI Global Union’s “Top 10 principles for Ethical Artificial Intelligence” provides concrete demands regarding its transparency and application in the workplace. UNI has been vocal about the ‘urgency of now’ regarding Artificial Intelligence and ensuring that workers share in the benefits of AI, rather than becoming its victims.