Humanists helping to shape global AI regulations

Humanists UK and Humanists International are currently engaged in work alongside civil organisations to positively shape European Union (EU) and worldwide regulations around artificial intelligence. 

Drawing from the Luxembourg Declaration on Artificial Intelligence and Human Values, adopted by members of Humanists International last year, humanists around the world are working to make sure that the development of artificial intelligence remains aligned to human needs, human wellbeing, and the protection of vulnerable people.

European AI regulations

On Monday 8 June, philosopher Dr Dorothea Winter of the German Humanist Association and the Humanistic University of Berlin engaged in a high-level meeting with religious leaders, civil society, philosophical organisations, and MEPs in Brussels.

On 9 June, Humanists UK’s own Director of Advancing Humanism, Liam Whitton, one of the principal authors of the Luxembourg Declaration, spoke at the European Parliament as part of an Article 17 dialogue session on how best to regulate the use of AI in the context of medicine, the loneliness epidemic, elder care, digital safety, and issues faced by young people.

Liam’s remarks focused on three recommendations for Members of the European Parliament ahead of the Digital Omnibus on AI coming to a final vote, drawing from the Luxembourg Declaration. 

Specifically, Liam asked MEPs to recognise the legislative gap around ‘digital companion apps’ marketed to lonely people, which have been linked to suicide attempts and mental health deterioration. He also raised concerns about delaying the 2024 AI Act’s rules on ‘high-risk rules’ for up to 26 months, in which time under-regulated AI systems could become integrated in hospitals, as in the US, impacting staffing, diagnostic, triage, and care decisions. 

His final point was on the proposed watering down of EU requirements for AI companies invest in AI literacy in a time of rising online misinformation. Liam urged MEPs not to do this, and instead to work with the big tech companies to come up with a European strategy for AI, health, and digital literacy, which could in turn help to make the law’s AI literacy requirements both measurable and enforceable.

Humanists UK had the closing comment of the session, recommending to European Parliament Vice President Sberna that the fruitful Article 17 session be continued in an ongoing format, recommending the model established by the UK’s AI Faith & Civil Society Commission established during Rishi Sunak’s premiership.

Humanists UK is represented on the AI Faith & Civil Society Commission through its patron, AI expert Professor Kate Devlin. Devlin also advised on the drafting of the Luxembourg Declaration.

Human-centric care: more than the right words in the right order

The loneliness epidemic in Europe was a recurring theme in the debate. In his points addressed to the panel on regulating AI companion apps, Liam offered a critical distinction between the capabilities of even sophisticated large language models and real human beings.

Warning that the embedding of insufficiently AI companion apps into healthcare risked ‘corroding our standards for human-centric care’, Liam drew example on the experiences of humanist pastoral carers in the UK and the Netherlands and said:

‘Humanists in Europe have been providing emotional support and care to patients for over 80 years, and in that time, humanist pastoral carers and chaplains have learned a lot about the subtleties and difficulties of providing emotional support to people facing health concerns. Body language, tone, subtext, humour. There is an art to listening.’

Later in his speech, he reflected on the words of Humanists UK Vice President, philosopher Bertrand Russell, who said that ‘The good life is one inspired by love and guided by knowledge.’

He concluded:

‘AI will, however imperfectly, no doubt continue to aid in the expansion of human knowledge. It certainly has a role to play in the enrichment of human health and welfare. But AI cannot supply love. It can simulate, but never embody, human empathy, human compassion, or human kinship. 

‘These are human traits hardwired into us by our very survival as a species. Our evolution. We only exist today because, for all of our faults, human beings learned to love one another and care for each other. 

‘Fairness by design must acknowledge the interconnectedness of health and happiness for every human being. This inevitably means greater caution where caution is the alternative to allowing harm.’

Convergence of values

A striking feature of these recent sessions at the European Parliament was the remarkable common ground across religious and non-religious groups. 

For example, the European Parliament session featured speeches from Humanists UK, the Catholic Church, the Chief Rabbi of Belgium, Buddhist and Muslim leaders, and secular global NGOs like UNICEF and Eurochild. Most speakers, whether religious, humanist, or undeclared, evinced a strong humanistic rhetoric, with words like human-centric repeating frequently and appeals for AI to serve humanity, align with human interests, and assist but not displace humanist reason, and not to undermine human rights or human dignity.

Liam summarised by saying:

‘The thing I’ve observed is how much convergence we’ve seen here today. We start with different sets of values and yet we’ve arrived in the same place: a very human-centric approach to AI.’

In parallel with these events in Brussels, the House of Lords recently held a debate on artificial intelligence instigated by the Archbishop of Canterbury, in which Lord Cashman recommended to the UK Government the example of the Luxembourg Declaration as a robust practical and ethical guide. The same debate also evinced a theme of convergence, with another peer, Lord Clement-Jones, described as ‘a convergence of values that goes well beyond any single set of beliefs’.