As I was just typing up my AI and human rights remarks from Copenhagen (more on that in my upcoming newsletter!), it was fantastic to see the UN’s Independent International Scientific Panel on Artificial Intelligence publish its first report.
The Panel was established by the UN General Assembly in 2025 as an independent, non-political scientific body to help build a shared understanding of AI and inform policymakers.
I’m all for it. 😊
The report provides an evidence-based look at AI’s capabilities, opportunities and risks. You can read our full summary below, but here are a few takeaways for now:
🤖 AI is fundamentally different from previous technological revolutions.
This isn’t just another emerging technology.
Unlike previous waves of automation that transformed physical labour, AI is the first technology capable of transforming cognitive and creative work at scale — from legal analysis and programming to medical diagnostics, scientific discovery and writing.
And… the speed is unlike anything we’ve known. Adoption has happened in months, not years…
⚖️ AI is now firmly a business and human rights issue.
The report makes clear that AI is changing how people experience many of their fundamental rights. It brings enormous opportunities, but also new risks around privacy, discrimination, surveillance, access to remedy and children’s rights.
Any business deploying AI needs to consider how people’s rights could be affected.
🔍 AI both create new inequalities — and amplifies existing ones.
The report highlights that AI often has the greatest impact on those who are already most vulnerable, including children, women, racial minorities and communities across the Global South.
As with so many business and human rights issues, existing inequalities don’t disappear with new technology. They can become even more pronounced.
🛠️ The good news? Businesses don’t need to start from scratch.
One of my favourite messages from the report is that many of the tools companies need already exist.
Human rights due diligence, impact assessments, transparency, accountability and rights-by-design are all highlighted as practical ways to identify and address risks to rights throughout the AI life cycle.
That was also the central message of my remarks in Copenhagen: we don’t need a new playbook—we need to apply the one we already have.
🌍 The challenge is keeping governance in step with innovation.
AI capabilities are advancing faster than regulation can keep up. The report describes this as an “evidence dilemma”: policymakers need evidence to make informed decisions, but by the time that evidence is available, the technology may already have moved on.
Top concluding takeaway: the greatest risks do not come from AI itself, but from deploying it at scale without appropriate safeguards, transparency and meaningful human oversight.
Some of us are designing these systems. Some of us are deploying them. Some of us are using them. And some of us are regulating them.
We all have a role to play in ensuring AI is developed and used in ways that respect people’s rights. 💪
Anna
PS: My update on the new Chinese measures is available on LinkedIn here. We’ve circulated our takeaways from our London just transition event to participants and we’ll be circulating them to you shortly as well.
