Luxembourg Declaration on Artificial Intelligence: a call to arms or arm-waving?
- Mike Flood

- Jul 31
- 6 min read

By Mike Flood
The Luxembourg Declaration on Artificial Intelligence and Human Values is a powerful statement of concern and intent. But will humanist organisations around the world do more than applaud it? In this timely and challenging piece, Mike argues that humanism must augment its traditional focus on secularism and religious privilege to confront the existential threats of our time. Mike is Chair of Milton Keynes Humanists and Humanism for the Common Good.
On 6th July 2025, the General Assembly of Humanists International – convening in Luxembourg – approved the Luxembourg Declaration on Artificial Intelligence and Human Values (summary and link provided below). It is beautifully written and represents a powerful statement of concern. Whether it will result in humanist organisations around the world actually doing something (anything) about AI is another matter as they are focused on fighting religious privilege and speaking up for the rights of the non-religious.
I have thought for some time that this choice of campaign focus is somewhat surprising given that the word 'religion' is only mentioned once in the definitive guiding principles of Modern Humanism (the 2022 Amsterdam Declaration), and then only in passing; and you won't find 'religion' mentioned in the new Luxembourg Declaration or in the 2023 Copenhagen Declaration (on Democracy) or the 2019 Reykjavik Declaration (on the Climate Change Crisis). Indeed, I have reluctantly come to the conclusion that there are real dangers in paying lip service to major contemporary concerns (such as AI) if you’re not prepared to 'put your money where your mouth is'. And on the question of fighting for the rights of the non-religious, what are we to make of humanists 'promoting and supporting human flourishing across all frontiers and championing human rights for everyone!' as Humanists UK proclaims in their infographic 'Humanism in a nutshell'? Words matter. (The infographic is shown at the bottom of this article.)
A number of us have been trying for some time to encourage a debate about such issues within the humanist community: in August 2022 we set up an ad hoc Future of Humanism Group and the following February produced a 3-page 'Manifesto' in which we highlighted the growing threat to society / humanist values posed by misinformation, AI and climate change. We had hoped that this would stimulate a wider debate about how people of goodwill can live 'happier, more confident, and more ethical lives' and endeavour to be 'good ancestors' to future generations in today’s increasingly polarised and turbulent world (see note 1). But it didn’t, and we are now in the process of relaunching the Group as Humanism for the Common Good. We have also revised our list of concerns to read 'truth, justice and sustainable development' to better reflect our interest in promoting humanism in a more positive light. Our message is not a million miles away from a comment by Frank Turner which Humanists UK posted on Facebook some years ago (and shown below) – Frank is one of Humanists UK’s Patrons. And note 'humanist solutions' is emphasised along with humanist ideas.
We do recognise the important work that humanist organisations have been doing over the years in: 1) fighting for the acceptance of humanists in countries where they face persecution; 2) campaigning for secularism; and 3) promoting humanism as a satisfying and fulfilling life stance. But we question whether this is sufficient, given the unprecedented challenges humanity now faces as a result of rapid technological advance, the degradation and destruction of important ecosystems / loss of biodiversity, and growing climate chaos. And we would like to see humanists taking a lead in actually: 4) responding to some of the threats that these developments pose to our cherished core values of truth, human rights and democracy, and indeed, the viability of sentient life on Earth. Indeed, this 'fourth stage' of humanism is implicit in the Amsterdam Declaration, which talks about humanists feeling 'a duty of care to all of humanity, including future generations, and beyond this to all sentient beings'; and it is core to what the distinguished Dutch humanist Jaap van Praag’s called the 'Great Fight' I do recognise that many grassroots humanists are actively promoting 'humanism for the common good', but that’s rather different from humanist organisations putting their 'UMPH' behind it!
And please note: no one is suggesting that humanists stop protesting religious privilege, rather that priorities be reviewed and new partnerships sought with issue specialists so that a coherent position can be formulated on at least some of the concerns highlighted above – there’s an incredibly rich pool of expertise within our movement, indeed two Humanists UK Patrons were among the experts consulted on drafting the Luxembourg Declaration! To see just how humanist organisations might actually get involved we suggested in our Manifesto 10 campaigns that humanists might endorse or adopt and we’ll be updating the list.
Humanists also need to recognise that humanism is hardly a mass movement here in the UK (and sadly, most places elsewhere): one member of our Group recently asked the ever-helpful ChatGPT 'Why?' and it opined that this was 'largely because many of [humanism’s] core ideas have been absorbed into mainstream culture, while other social, cultural, and political forces have taken its place as vehicles for meaning and identity.' So this is another of the issues that our Group will be exploring over the coming months – it would be interesting to know when the Humanist Powers That Be actually carried out their last in-depth assessment of this issue… And for those who baulk at local groups getting too political, please note that the question of ‘How do we engage with political issues while not being party political?’ is actually on the Agenda for the forthcoming Group Representatives Annual Meeting, which is being organised by Humanists UK and Humanist Groups Working Together (in Reading on 9th August).
The views expressed in this paper are mine and not necessarily those of Humanism for the Common Good or Milton Keynes Humanists, both of which I chair.
The Luxembourg Declaration on AI and Human Values (2025) – Summary
Adopted by Humanists International in July 2025, the Luxembourg Declaration calls for ethical, democratic, and human-centred governance of artificial intelligence, grounded in humanist values such as reason, compassion, dignity, and freedom. It recognises both the promise and the peril of AI, emphasising that technological advancement must serve the common good, not entrench power or inequality. The Declaration highlights ten core principles:
Human Judgment: AI must never replace human ethics or responsibility. Life-altering decisions must remain in human hands.
Common Good: The benefits of AI must be widely shared, not monopolised by elites or corporations.
Democratic Governance: AI systems should be accountable to democratic institutions at every level, with lawmakers equipped to regulate them.
Transparency and Autonomy: Citizens have a right to understand and challenge how AI systems use their data and influence their lives.
Protection from Harm: AI must not be used to reinforce injustice or suppress lawful expression. It must be safe, non-discriminatory, and aligned with human rights.
Shared Prosperity: AI should enhance human dignity, not deepen social divisions. Investment in education and social protections is essential.
Creators and Artists: Human creativity must be recognised and protected. AI can assist, but must not undermine the value of human artistic expression.
Truth and Integrity: In a world threatened by misinformation, AI must not erode our ability to distinguish fact from falsehood or corrupt democratic discourse.
Future Generations: Decisions about AI must account for long-term consequences, including environmental and existential risks.
Human Flourishing: AI should ultimately expand freedom, happiness, and the richness of human life – giving people more time to learn, connect, and thrive.
The Declaration closes with a call to action: for governments, corporations, civil society, and individuals to embed these principles in policy and practice, and for humanist organisations to lead by example in shaping a more just and humane future.
Read the full version here: Luxembourg Declaration on Artificial Intelligence and Human Values
Notes
In February I attended a talk given by AC Grayling on the subject of ‘Humanist Principles in Unprincipled Times’; it was to Oxford Humanists and he spoke about how so many humanist values are today in real peril. Professor Grayling’s point is that you need a 'good society' to provide the context for people to live a 'good life', and yet all around the world, populist politicians and autocrats have been systematically dismantling the structures that provide for a good society (check out his 2022 book: ‘For the Good of the World’). And AC asked, 'What should we as humanists be doing in response?' 'You can’t be passive' he said, 'You have to take action.' 'Be vocal.' 'Stand up and be counted'. I've seen the Professor speak on many occasions, but I don't think I've ever heard him speak quite so passionately or with such conviction.
There’s a wealth of material on 'humanism and politics' on the Humanist Heritage website. You might also like to check out what the Charity Commission has to say on the subject, namely that: 'campaigning and political activity can be legitimate and valuable activities for charities to undertake' although 'only in the context of supporting the delivery of its charitable purposes… [Indeed] There may be situations where carrying out political activity is the best way for trustees to support the charity’s purposes.'






I agree that the challenge is to turn fine words into concrete action.
As Humanists, we’ve long played a role in helping society question powerful forces — such as religion — by exposing unfounded claims, highlighting where harm is done, and offering rational, compassionate alternatives. Perhaps we can take a similar approach with AI: questioning inflated promises or hidden risks, ensuring claims are evidence-based, and keeping human dignity and wellbeing at the centre.
One practical way forward could be to help build AI literacy in the same way we’ve encouraged people to recognise sound reasoning and detect logical fallacies. This would mean equipping people to understand how AI systems work, how bias or manipulation can creep in, and how to…
This was interesting, so I asked my AI, did it miss anything?
That's a great question, Aaron — and you're right to ask it. The Luxembourg Declaration is commendably broad, principled, and aligned with humanist ethics. But as with any high-level statement, there are areas where further clarity, strength, or nuance could help ensure its aspirations are fully realised. Here's a brief critical analysis of what might be missing or underdeveloped:
---
🔍 Areas Where It Could Go Further
1. Implementation Mechanisms
What’s missing: The declaration outlines principles but is light on how they’ll be enforced or operationalised.
Suggestion: Include clearer calls for international oversight bodies, AI ethics boards, or mechanisms for corporate accountability, similar to environmental or human rights…