What we do
We’re building an open community that reads, researches and advises on advanced AI, so that India understands the technology and has a real voice in how it is governed.
From understanding the problem to informing the people who decide.
Reading groups
Small groups that meet at and around IIT Delhi to read one paper or report together and talk it through. Open to anyone, whatever they study or do.
AI Safety Fellowship
Six weeks, online, in a small group. Each week takes one question about how advanced AI can go wrong and what people are doing about it.
IASN Talks
Talks by people working on AI safety, with a recording or a write-up after each one.
Research
Mentored projects in alignment, interpretability, evaluations, governance and the economics of AI, written up in public so others can build on them.
Policy
Clear, independent analysis for the people making decisions about AI in India: government, regulators, companies and civil society.
Incident register
A public record of cases in India where AI systems harmed people or failed, each with its sources, and a way for anyone to report a new one.
How we plan to get from first conversations to better decisions.
Gather
Bring together people from every field who care about how advanced AI goes.
Learn
Reading groups and workshops that build real understanding of the problem.
Research
Mentored projects that produce work researchers and policymakers can use.
Advise
Put clear, independent analysis in front of the people who decide.
New to AI safety? Start with the reading list.
It starts with what AI alignment means and works up to current research on training, oversight and the insides of language models. You can begin with no background: the maths and machine learning the later papers need are listed with it. It’s open to everyone, no form required.