MATS Research

80

5.3K

MATS Research retweetledi

Ryan Kidd@ryan_kidd44·3 Tem

Expressions of interest to mentor for MATS Winter 2027 Program close soon! If you work in the fields of AI alignment, governance, interp, or security and want to mentor with MATS, please let us know by Jul 10! matsprogram.org/mentor-eoi

English

3

26

2K

MATS Research@MATSprogram·30 Haz

Join MATS in Seoul for our evening social mixer at ICML on Day 2 of the main conference! Meet and talk to research managers, MATS alumni, and conference attendees about the most important questions in AI alignment and security. RSVP here: luma.com/8f6i91q9

English

2

65

4.6K

MATS Research@MATSprogram·29 Haz

Noticing what's already strange, imagining what may come next, and making the invisible workings of AI beautifully apparent: each pushes the frontier forward. Read it here 👉 matsresearch.substack.com/p/creativity-i…

English

1

15

909

MATS Research@MATSprogram·29 Haz

🧬 Michael Yu's VFUSE A mechanistic interpretability approach surfacing the protein-design features that activate only on hazardous structures.

English

15

1.1K

MATS Research@MATSprogram·29 Haz

What happens when you stop trying to make AI models do what you want, and start paying attention to what they actually are?Our latest post explores creativity, imagination, and beauty in AI safety through the work of our fellows 🧵

English

2

5

122

11.4K

MATS Research@MATSprogram·8 Haz

🚨 Applications for MATS Autumn 2026 close tonight (June 7 AoE)! Spend 10 weeks fully funded working with mentors from Anthropic, DeepMind, OpenAI, Redwood Research, SecureBio, and more. New this cohort: 🧬 Biosecurity 🚀 Founding & Field-Building Apply now: matsprogram.org/apply

English

6

11

156

60.1K

MATS Research retweetledi

Luke Drago@luke_drago_·5 Haz

Applications for @MATSprogram close in two days! Learn more about our stream and apply today (link in replies)

Luke Drago@luke_drago_

If you want to research interventions to gradual disempowerment or the intelligence curse, @LRudL_ and I are mentoring a @MATSprogram stream this autumn. Many people have asked me “what’s the plan to make this go well?” Right now, there’s not one. You should help fix that. 🧵

English

4

35

5.4K

MATS Research retweetledi

Akshat Naik@aksh_n0·4 Haz

Trained monitors can be strong low-cost alternatives to prompted frontier models for black-box scheming detection. Our fine-tuned open-weight monitors detect scheming/sabotage in agent trajectories better than small prompted models and are on the cost-performance frontier. (1/n)

English

7

29

6.9K

MATS Research retweetledi

Celeste (ICML!)@celestepoasts·4 Haz

New research from @japhba and I! Activation Oracles are a pretty cool interpretability tool. They answer natural questions about activations, but they suffer from vagueness and hallucinations. Can AO training be improved? Turns out: Yes! We identify four fixes that make AOs substantially more useful!

English

12

25

227

32.4K

MATS Research retweetledi

Cas (Stephen Casper)@StephenLCasper·29 May

Apply by June 7 to work with MATS in the Fall. In the MATS stream that I coordinate, we will work on impact-oriented technical AI governance research, potentially including research on open-weight models, AI safeguards, AI incidents, technically rigorous AI policy, etc.

1/ 🚨 MATS Autumn 2026 applications are now open. 10-week fully-funded fellowship for aspiring AI alignment, security & governance researchers and field-builders. 📍 Berkeley + London 📅 Sep 28 – Dec 4, 2026 💰 $5000/month stipend + $8,000/month compute Apply by June 7 AoE ↓

English

7

117

9.1K

MATS Research retweetledi

Nick Raushenbush@nickraushenbush·30 May

Excited to serve as a mentor on the @MATSprogram founding & field-building track this autumn. If you're interested in moving into AI security and alignment, critical cyber security, or biosecurity, you should consider applying. Deadline is June 7th.

🚨 New for MATS Autumn 2026: the Founding & Field-Building track. A fully-funded track for founders, field-builders and amplifiers ready to launch and scale new AI safety initiatives. Apply by June 7 AoE ↓

English

1

834

MATS Research retweetledi

Nora Ammann@AmmannNora·31 May

Apply by June 7th to work with me on the institutional stack for the post-AGI world: moral reasoning, bargaining, structured transparency, trust infrastructure, and resilience capabilities.

🚨 New for MATS Autumn 2026: the Founding & Field-Building track. A fully-funded track for founders, field-builders and amplifiers ready to launch and scale new AI safety initiatives. Apply by June 7 AoE ↓

English

10

115

11.7K

MATS Research retweetledi

Max Nadeau@MaxNadeau_·2 Haz

This track—and MATS more broadly—is a great way to develop an idea for a new AI safety org to the point where it's ready for CG funding. I bet we'll end up making some seven- or eight-figure grants to new orgs that come out of MATS this year.

🚨 New for MATS Autumn 2026: the Founding & Field-Building track. A fully-funded track for founders, field-builders and amplifiers ready to launch and scale new AI safety initiatives. Apply by June 7 AoE ↓

English

4

101

7.8K

MATS Research retweetledi

Charlie Petty@incredutility·2 Haz

Geoff Ralston @gralston and I are teaming up to mentor a biosecurity stream for @MATSprogram this fall. Check out the other great streams!

Streams led by: · @dewierwan_ (BlueDot) · @MikeMcCormick_ & @nickraushenbush (@HalcyonFutures ) · @gralston & @incredutility (Bio Action Plan) · @LisaThiergart & @luiscosio (@SL5TaskForce) · @AmmannNora (ARIA) · @RosieCampbell (@eleosai ) · @degerturann (@metaculus)

English

3

11

1.6K

MATS Research retweetledi

Dillon Plunkett@dillonplunkett·3 Haz

I’m mentoring Autumn 2026 @MATSprogram Fellows interested in doing AI welfare research. The application deadline is this Sunday (6/7). More info in this thread: