Community

Events and speaker sessions

AI Safety India runs free AI safety talks, reading groups, research sprints and meetups in Delhi NCR, Bengaluru, Agartala and online.

Subscribe on LumaJoin the WhatsApp community
The room watching Ojasvi Gupta’s introduction slide at her talk in Bengaluru, October 2026.
Arunachal PradeshAssamChandigarhKarnatakaManipurMeghalayaMizoramNagalandPunjabRajasthanSikkimTripuraUttarakhandTelanganaBiharKeralaMadhya PradeshAndaman & NicobarGujaratLakshadweepOdishaDadra and Nagar Haveli and Daman and DiuJammu & KashmirChhattisgarhDelhiGoaHaryanaHimachal PradeshJharkhandTamil NaduUttar PradeshWest BengalAndhra PradeshPuducherryMaharashtraLadakhAI Safety NITAAgartalaDelhi AI Alignment and Safety CollectiveDelhiAI Safety India BengaluruBengaluru
Boundaries traced from Survey of India maps by DataMeet (CC BY 4.0). Not to scale.

What’s on

Upcoming

Next event, in 2 days

AI Safety India Spotlight: How I Got Into AI Safety Research, with Pratinav Seth

Pratinav Seth leads AI safety and alignment research at Lexsi Labs.

Sunday, 15 November, 7:00 to 8:00 pm ISTOnline
RSVP on Luma: AI Safety India Spotlight: How I Got Into AI Safety Research, with Pratinav SethAdd to calendar: AI Safety India Spotlight: How I Got Into AI Safety Research, with Pratinav SethShare: AI Safety India Spotlight: How I Got Into AI Safety Research, with Pratinav Seth

Past

From the room

The talk that opened the AI Safety India meetup in New Delhi.
Meetup · New Delhi · September 2026The first Delhi meetup opens
Ojasvi Gupta presenting her talk on bias and fairness in AI in Bengaluru.
Talk · Bengaluru · October 2026Ojasvi Gupta on bias and fairness in AI
A small group talking in a circle of chairs after the talks at the AI Safety India meetup in New Delhi, September 2026.
Meetup · New Delhi · September 2026Small groups after the talks
Attendees chatting across their desks between talks at the AI Safety India meetup in Bengaluru, September 2026.
Meetup · Bengaluru · September 2026Conversations between talks

Write-ups

All posts on Substack
Speaker session · Hackathon winners · 5 August 2026

Do multilingual vision-language models abstain under cross-modal conflict in low-resource languages?

Anvesh Reddy Lankala, Vicky Feliren and Akansh Jain, winners of the Apart Research Global South AI Safety Hackathon (Asia-Pacific track)

What they found
  • VLMs get about twice as text-biased in Hindi and Telugu as in English.
  • High abstention can be a mask: Qwen followed the text 81% of the time when it failed to use the image.
  • Steering vectors trained on English transfer for abstention models, but fail for heavily text-biased ones.
Read the write-up on Substack

Hear when the next session goes up.

Subscribe on Luma

Programmes

Hubs