6th November 2025
Artificial intelligence may feel like the latest leap into the unknown, but as venture capitalist and philanthropist Saul Klein reminded guests at Prism’s final 20th anniversary event, “whatever you think AI can do today – it’s the worst it’s ever going to be.”
The evening opened with Rhodri Davies, founder of Why Philanthropy Matters, who offered a clear framework for understanding how AI is already shaping the charity world. AI, he explained, is influencing philanthropy on three fronts at once: through new “tech for good” solutions tackling social issues, through major changes in how charities operate day-to-day, and by reshaping the very nature of the challenges they exist to solve.
Far from a new phenomenon, Saul explained, AI is part of a long tradition of human invention – one that stretches back to Alan Turing in the 1940s beyond. It’s most recent stops including the birth of the internet, cloud computing and apps.
The difference now is that, for the first time, “AI is in our hands. It was done to us for years and now we have the chance to do AI with us.” We now have a chance to guide it and develop it in ways it can benefit society.
Reimagining innovation with values
Saul, who co-founded Phoenix Court and serves on the Prime Minister’s Council for Science and Technology, is used to looking decades ahead. His venture funds invest in what he calls “practical science fiction” ideas that seem outlandish now but could transform lives in 10 or 20 years.
Yet his story was firmly rooted in the present. Phoenix Court’s foundation is guided by a deceptively simple mission: to be a good long-term neighbour and help people realise their full potential.
Ten percent of Phoenix Court’s profits are given to its foundation, and 80 percent of its grants stay within a mile of their King’s Cross office – one of London’s most deprived areas. The team hosts quarterly Neighbourhood Labs, bringing together local organisations, schools, and major institutions such as the British Library and The Alan Turing Institute.
Among the most popular are the AI Labs, designed to help local people and charities understand and use AI tools safely. “AI literacy is absolutely critical now,” Saul said. “People shouldn’t be afraid of it – this is the first time it’s for us.”
His message was clear: innovation without values isn’t innovation worth pursuing. True progress, he argued, starts with neighbours learning together how to harness technology for good.
A legal lens on responsible AI
Providing the legal perspective was Alan Baker, AI Partner at Farrer & Co, who advises clients on everything from data protection and intellectual property to the ethics of AI governance.
While the technology can feel overwhelming, Alan reassured the room that “the legal framework in the UK is familiar.” Rather than sweeping new laws, the government has chosen a “pro-innovation, sector-based” approach, applying principles that charities already know from GDPR such as accountability and transparency.
His advice to philanthropists and charity leaders was practical:
- Treat AI as you would any other regulated process. If you can run a data protection impact assessment, you can run an AI risk assessment.
- Keep records of how and why you use AI tools.
- Be open with staff and beneficiaries – transparency builds trust.
- Involve everyone. “Get everybody talking,” he said. “There’s value in the creative genius of an organisation when everyone is involved.”
Alan illustrated the final point using the example Farrer’s AI group which brings together voices from across the firm to make sure tools are relevant across different departments, challenge assumptions and shape responsible use.
Rethinking philanthropy’s role
The discussion turned to what philanthropy can do that the market cannot. Saul argued that issues such as AI safety, bias, and accountability are classic market failures, areas where commercial incentives fall short and philanthropy must step in.
“Where are the philanthropists funding AI safety?” he asked. “Most models we use today are built on data that’s biased or even stolen. Philanthropy can help build the alternatives.”
One of the biggest ethical challenges is the issue of pitting ‘machine against machine’. Today, millions of applications and assessments, from insurance to bank loans. are made using AI. As Saul said, “Depending on how the machine is trained and the biases that are inherent in those models… this is the big problem.”
Then there is the challenge that the AI platforms most widely used in the UK are built and controlled in the US. Saul urged funders and organisations to work with communities and partners across the UK, Europe and Asia to help create credible alternatives.
He challenged the assumption that AI will erase jobs or human connection:
“It’s not going to replace human work — it’s going to replace boring human work. It’s going to allow us to think, collaborate, and create.”
Both speakers agreed that philanthropy’s strength lies in its humanity. Technology may speed up grantmaking and improve access, but “people give to people,” as Prism’s CEO Anna Josse reminded the audience. “That element will always be there.”
The environmental cost and opportunity
Questions from the audience turned to AI’s heavy energy and water use, now estimated to rival aviation’s carbon footprint. Saul acknowledged the challenge:
“AI systems are massive users of energy. Many Bitcoin mining companies have now pivoted to become AI infrastructure businesses.”
But he pointed to emerging hope: new low-energy computing methods and genuine carbon-removal technologies that could make AI more sustainable. “Carbon offsets do nothing,” he said. “Carbon removal actually removes carbon.”
When asked about the carbon footprint of AI, Alan explained, “Taxing data centres won’t work-they will just move”
Many data centres are now moving to colder countries in a move that could be seen as both positive and negative – with the risk of increased warming in these locations. But he reminded everyone that the big tech companies have a huge financial incentive to solve the energy issues which will in turn relieve pressure on the environment.
Alan said, “The real solutions to environmental factors will come from technology, not regulatory punishment.”
Harness the power of AI: Practical steps for philanthropists and advisers
- Start small and local. Test AI tools that make everyday processes easier, from drafting applications to analysing impact, but keep human oversight at the core.
- Stay transparent. Document how you use AI and be open with your teams, donors, and beneficiaries.
- Champion AI literacy. Share learning with partners and funders as Phoenix Court’s Neighbourhood Labs do.
- Fund ethical innovation. Support research and projects that build fair, safe, and environmentally sustainable AI.
- Keep people first. As Prism’s Anna Josse reminded everyone, technology can assist philanthropy but never replace its human heart.
Looking ahead
From the legal frameworks already in place to the neighbourhood initiatives building confidence in new tools, the evening painted a picture of both realism and optimism.
AI is not the future – it’s already here. The question for philanthropists and their advisers is how to ensure it serves humanity rather than the other way around.
As Saul put it, “If all we’re going to do is take what we’re given, we’ll get what we’re given.”
As Prism celebrates securing over $1 billion in donations for good causes around the world in 2025, Anna closed by saying, “As Prism moves into its 21st year, we have ambitions to raise another $1b of donation income and no doubt AI will help get us there”.
Thank you to our 20th Anniversary sponsors whose support has helped us to bring people together and explore some of the big issues in Philanthropy. These include, Charles Stanley, Christie’s, Edmond De Rothschild, Evelyn Partners, Farrer & Co, Maurice, Turnor & Gardner and Bluston Charitable Trust.
