The Paris AI Action Summit: Steph Wright’s Thoughts
The Grand Palais, Paris
Head of Scottish AI Alliance Steph Wright attended the Paris AI Action Summit on 10 February - here are her thoughts from the day!
Earlier this week, I had the absolute honour of being invited to participate in the Paris AI Action Summit and before all the conversations fade in my memory, I thought I would do a wee brain dump into this blog so buckle up and here we go, it’s a long ‘un!
The Parisian weather was what can only be described as ‘dreich’ on the Monday morning as I set off early to get into the Summit. I was expecting a huge queue to go through security but I guess it being 08.15 meant it wasn’t actually particularly busy. Picked up my badge and headed into the beautiful Grand Palais. The venue was huge and suitably grand as its name implied and it had a great range of projects from all over the world exhibiting AI products, projects and initiatives. Creative works were also projected onto a central area which looked fantastic. It wasn’t long before I bumped into people I knew which was a huge relief and we all shared a curiosity as to how the day of conversations will go.
Kicking off proceedings was Anne Bouverot, the Special Envoy to the Paris AI Action Summit followed by an opening keynote from the renowned Dr Fei Fe Li, the founder of ImageNet. I am huge fan of Dr Li’s work (I loved her book The Worlds I See) but was rather baffled at the content of her talk which gave a general overview of the history of AI. Of course acknowledging that it was what she was asked to speak on, I did wonder (along with a few others) that surely this audience would already know all this? She did end with a call for human-centred AI, for science and not science fiction, and for AI that is for good.
The vastness of the venue made it somewhat difficult at the start to figure where to go for the various parallel sessions. After going to the completely wrong side of the building and walking through somewhere I definitely don’t think I should have been in, I finally got to the Auditorium where I had hoped to join a session about Building Trustworthy AI. However, my adventures getting lost meant I got there too late and it was full (I did register for the session but that didn’t seem to matter). Och well, I thought and wandered over to the Main Stage where there was a conversation about Harnessing AI for the Future of Work and I was glad I did. Having supported the Institute of Work’s launch of the Pissarides Review a couple of weeks ago, the topic of the impact on AI on the future of work was at the forefront of my mind. There were lots of interesting points made in this session including:
- We need to give workers a voice
- Companies should be incentivised to keep their workforce as they roll out technologies like AI
- Technology is not often on the table in collective bargaining situations
- An inclusive transition is possible but not guaranteed
- We need ground rules set in people centred policies
- Concern about the intensification of work
- A good question asked – who pays for upskilling and reskilling?
But the speaker who resonated with me most (to my surprise as I am not at all religious) was Archbishop Paul Richard Gallagher from the Vatican who injected humanity into the discussions by reminding everyone that AI is not going to solve the problems of this world and that is an accelerator of inequalities and economic inequalities go hand in hand with political inequalities. He made the point that although work is important as the centre of economic activity, we need to not definite human value, dignity and worth by economic performance – that people are not commodities, that they are people with full dignity and full worth and that they must have their voice represented in the transition. He also said that we need to consider that work should not just be considered solely as an economic activity and that we need to look at how we can maintain solidarity in a world marked by the original sin of competition and meritocracy and ensure that we don’t exacerbate those tendencies and everyone’s rights and dignity is respected. This passionate plea was met with a big round of applause. During the mention of education later, he pointed out that education is not just a technical achievement, it is a human achievement and that purely looking at it from a technical perspective is the wrong way. He pleaded that we should look at education as a human endeavour and how can we enable the formation of people who can ask the bigger questions, people who can harness the power of AI to ensure a just future of all, and how do we assure them that they have dignity, worth and value beyond academic or economic achievement. This was powerful stuff and it was great to hear it on the first session on the main stage. I was so moved I felt the need to go and meet the Archbishop and thank you for his contribution.
A digital display a the Paris AI Action Summit 2025
Energised by this, I stayed for the next session about Creating a Virtuous Circle between AI, Creation & Information. Another esteemed panel of speakers including the President of the Republic of Estonia. Prof Jane Ginsburgh from Columbia University reminded everyone why we have copyright law in the first place… it’s there to protect and preserve creative endeavour, not to be a barrier to innovation. There is a virtuous way of using creative outputs to train models by licensing and of course the non-virtuous way that seems to be the practice at the moment. The panel discussed initiatives that offer artists and creatives an “opt out” but it was noted that you can’t opt out if you don’t know your data is being used so transparency is needed. An impassioned speech from Nabil Ayouch, a film director, boldly stated that artists are not afraid of AI and that what was artists do has value. Paraphrasing Donald Trump’s recent “drill baby drill” comment, Nabil said “regulate baby, regulate” as regulatory frameworks enable innovation to thrive. Humanity and cultural diversity need to be at the core and another call for transparency in data usage; artists have the right to be informed that their work is being used, they have a right to fair compensation. Once again, a passionate, powerful call from a speaker.
Next up after lunch (which consisted of tiny canapes) was the workshop titled AI and Children’s Development. I was somewhat curious to understand why they called them workshops as the ones I attended involved no audience interaction and instead just had lots of quickfire input from a wide range of speakers. This session was no different, there were in fact about 15 speakers! A rather serendipitous meeting with Anne-Sophie Seret from everyone.ai at the Children’s AI Summit in London last week resulted in our Children and AI project being mentioned in the Beneficial AI for Children Coalition booklet which was given out at the Summit! I also managed to squeeze in some of our children’s Call to Action leaflets amongst the audience members. The session started with a clear statement that children are not mini-adults and shouldn’t be treated as such, and a desire that we want the next generation to be fully functional human beings! One of the speakers was the fantastic Nesta Quiroz-Aitken, there to represent the voices of the children at the Children’s AI Summit - “you write the laws but we bear the consequences”. Representatives from OpenAI and Google were part of the session but disappointingly left right after their contributions… I would have loved to have given them a flyer with the children’s calls to action. Ultimately a thought-provoking dialogue from a range of stakeholders (and I got to meet and chat with Stuart Russell) I look forward to exploring how we can contribute to the coalition for beneficial AI for children.
The Exploring Children’s Rights & AI project mentioned in the Beneficial AI for Children Coalition leaflet
Another workshop I attended was titled A Call to Rethink AI: Why Gender Equality and Human Rights Matter Now More than Ever. The session included the launch of a call for gender to be to be an integral part of the AI Action Summit led by the Global Partnership for Action on gender-based Online Harassment and Abuse and signed by the governments of Australia, Canada, Chile, Denmark, Finland, France, Iceland, Mexico, Republic of Korea, Spain, Sweden and the United Kingdom. The speakers stated that AI is a human rights issues and a gender lens needs to be mainstreamed and integrated into every stage of AI. It was brilliant to hear from the incredible Meredith Whittaker of Signal AI and the AI Now Institute. She called for a redefinition of “innovation” as its current definition is just a synonym for the broken system we have. It was also great to meet Caitlin Kraft-Buchanan from Women at the Table in person. Read more about their AI and Human Rights Toolkit in a blog she wrote for us last year.
Then I ran back to the Main Stage to catch the end of the session titled Bending the Arc of AI towards Public Interest: towards a Resilient and Open AI Ecosystem featuring the amazing Dr Abeba Birhane of Trinity College Dublin reminding everyone that Big Tech have reneged on almost every voluntary commitment around responsible AI and that they should embrace being challenged. Janet Haven from Data and Society called out the claim that AI will solve social issues (i.e. it doesn’t) and the brilliant Vilas Dhar from the Patrick McGovern Foundation saying that there is a virtue in building a world where we help each other with AI. Why are we investing in building bigger and better AI when we could instead invest in what we already have to help each other?
I think I might have shouted a few “woop woop”s at his comment. Why indeed, why instead are we striving to achieve AGI when there is no reason other than to see if we can?
The official programme of the day ended with a speech from President Macron himself announcing a €109 billion investment in French AI, saying we all need to collaborate and work together for our AI future. And boasting about France’s renewable energy supply, he joked there’s no need to “drill baby, drill”, just “plug baby plug”.
In the evening, I attended the welcoming event of the Summit from Make.org around AI and Democracy at the beautiful Maison de Chimie where they launched their Manifesto for a call for a Worldwide Alliance on AI and Democracy. Another great programme of speakers included Audrey Tang, former Digital Minister for Taiwan and Sviatlana Tsikhanouskaya, President-Elect of Belarus. We look forward to connecting with Make.org to explore any future collaborations!
Overall, it was a jam-packed day full of interesting conversations and meeting lots of interesting people. It was great to know that we are talking about the right things but there was a palpable tension in the discussions between those driven by narratives of economic growth and productivity vs those driven by social equity and mitigation of real and present harms. The environment was barely mentioned although I realise I wasn’t able to attend all sessions and the labour exploitation and resource extraction that powers AI barely got a look in. There was a general agreement however that the current concentration of power and influence is not sustainable nor is it for the benefit of humanity as a whole.
But what can we do about it? I am very excited about the announcement in Paris of Current AI, an almost €400M initiative focussed on “building a future where open, trustworthy technology serves the public interest.” But with the news that the US and UK have opted to not sign the declaration from the Summit (noted that it is more symbolic than action driven) which focuses on Inclusive and Sustainable Artificial Intelligence for People and the Planet, the divide between the schools of thought on AI is widening. But I firmly believe that economic growth and ethics and responsibility, sustainability and inclusion are not mutually exclusive concepts. Economic growth grounded and centred in ethics, responsibility and inclusion will ultimately be more sustainable and we are primed in Scotland for this having set trustworthy, ethical and inclusive AI at the core of our national strategy.
So, I would call for those of us who have a voice to use that voice to continue to champion a different narrative, a narrative in which AI benefits us all. In the words of Rachel Coldicott at our Scottish AI Summit in 2024, “let’s make AI work for 8 billion people and not 8 billionaires.”