Hub · NIT Agartala, TripuraOpen globally

AI Safety NITA

A research community for AI safety.

Join AI Safety NITASee meetups
Line drawing of Ujjayanta Palace in Agartala, reflected in its lake, with palm trees.
  • WeeklyPaper reading, every Saturday
  • Any levelNo prior AI safety knowledge needed
  • OnlineAnd in person at NIT Agartala
  • 4 areasOf research, multi-agent safety first
AI Safety NITAAI Safety NITA Join

Now and next

  1. Paper reading club

    Structured weekly discussions on core AI safety papers. Beginner-friendly and open to all.

    Active now

  2. Research working group

    Members from the reading club move into active research on open problems, with real output: papers, preprints and policy memos.

    Planned

Research focus

Multi-agent safety first: what fails when many AI systems interact.

  • Multi-agent safety

  • Alignment and control

  • Interpretability

  • AI governance

Who it’s for

  • Students

    CS, ECE, data science, any year. No prior AI safety knowledge needed.

  • Early researchers

    You do ML work and want to move toward safety problems.

  • Experienced researchers

    Join as a core collaborator or research mentor.

  • Mentors

    A strong background in technical alignment or policy.

  • Funders

    Help build a pipeline of AI safety researchers in India.

  • New to AI safety

    Start with a free six-week course, Technical AI Safety or AI Governance. Our AI safety career guide for India shows the next steps. Questions about the chapter? Write to us through the AI Safety India contact page.

Meetups in Agartala

In person, from our Luma calendar.

  1. Saturday14 Nov
    Research sprint

    AI x Epistemics Research Sprint

    5:30 pm ISTAgartala
    RSVP on Luma: AI x Epistemics Research Sprint
Attendees talking between sessions at an AI Safety India meetup in Bengaluru.

Bring curiosity. We will bring the paper.

Who runs it

A student chapter at NIT Agartala. AI Safety India itself began as a reading group here.

With the support of

  • AI Safety India

    Parent organisation: curriculum and mentors

  • Pathfinder Fellowship

    Kairos

Research focus

Multi-agent safety

Coordination failures, unsafe emergent behaviour, and runtime oversight in networks of AI agents.

When many AI systems interact and coordinate, new failure modes appear that single-model alignment does not address. It is one of the least studied problems in AI safety, and it is where we spend most of our time.

Research focus

Alignment and control

Scalable oversight, control evaluations and deceptive alignment.

Research focus

Interpretability

Mechanistic interpretability, probing and representation analysis.

Research focus

AI governance

Policy analysis, risk frameworks and regulatory research.

Join AI Safety NITA

Read with us this Saturday.

Step 1 of 2Saved

About you

Two short steps, about 3 minutes in all.

Session invites come here.

For example, NIT Agartala or BITS Pilani.

Current role

Programmes

Hubs