Joined August 2015
- Tweets 47
- Following 159
- Followers 154
- Likes 412
8 Photos and videos
Pinned Tweet
28 Sep 2025
🚨Paper Alert 🚨
Excited to release the Triplet adversarial defense at #EMNLP2025, a new approach extending circuit breaking (2024, by @andyzou_jiaming, @hendrycks et al.) which achieves < 5% ASR on embedding-level attacks! #AISafety
🔗 Link: arxiv.org/abs/2506.11938
🧵(1/6)
1
1
20
7,624
Samuel Simko retweeted
If you are a PhD student or early-career researcher, apply before the deadline: June 21, 2026. Aug 31 – Sep 11 · Preliminary speaker list and applications: lnkd.in/ekrPt4m9Amazingly, this is the 50th MLSS, and it coincides with the 25th anniversary of our lab at Max Planck (2/3).
1
24
223
54,260
Samuel Simko retweeted
Jun 4
Open-weight LLMs ship with safety training that can be stripped in a few hundred fine-tuning steps. Can current defenses stop this?
We built and open-sourced TamperBench, the first unified framework for evaluating tamper resistance, and the answer is mostly no. 1/7
1
14
28
4,409
Jun 2
I’m in the Bay Area for the rest of the week 👋
Would love to connect with AI safety researchers while I’m here. Also curious about any events happening this week.
1
7
981
May 10
I’ll be speaking at the AIxBio event in Zurich on May 13th at 18:00! Join for a discussion of what current AI systems can (or cannot) do and how their risks can be reduced.
Registration link: luma.com/a3nyvkkp
9
391
Samuel Simko retweeted
May 3
Excited for our #ICML2026 papers at @JinesisLab @MPI_IS @UofTCompSci @TorontoSRI @VectorInst! We present papers that advance the research frontiers of (1) Causal LLMs, (2) AI for Science (physics), (3) Multi-Agent LLMs via mechanism design, and (4) Adversarial Defense by honeypot. Congrats to all our student authors and collaborators, esp. @TerryJCZhang @SimkoSamuel @EmanuelTewolde @ivakshi_s @andrewkihyun @PepijnCobben @yahang_qi @FurkanDanismann @bschoelkopf and many others!🎉
11
76
3,884
May 2
Stage I for MARS V closes Sunday, 3 May at 23:59 AoE!
Apr 27
[Call for applicants]
My supervisor @ZhijingJin (UofT, CIFAR AI Chair) and I will be mentoring a project for MARS V, a part-time research programme for AI safety research.
MARS provides a one-week in-person kick-off in the UK, compute, and research management support! 🚀
The projects are:
🛡️ Adversarial defenses for LLMs using causal methods
🌐 Evaluating risks from AI-assisted authoritarianism
👉 Apply by May 3rd. Applications are reviewed on a rolling basis: caish.org/mars @CambridgeAISafetyHub
11
1,741
Apr 27
[Call for applicants]
My supervisor @ZhijingJin (UofT, CIFAR AI Chair) and I will be mentoring a project for MARS V, a part-time research programme for AI safety research.
MARS provides a one-week in-person kick-off in the UK, compute, and research management support! 🚀
The projects are:
🛡️ Adversarial defenses for LLMs using causal methods
🌐 Evaluating risks from AI-assisted authoritarianism
👉 Apply by May 3rd. Applications are reviewed on a rolling basis: caish.org/mars @CambridgeAISafetyHub
2
9
92
17,976
Samuel Simko retweeted
We'll be organizing the Machine Learning Summer School in Tübingen to be held Aug 31st-Sept 11th, featuring top speakers across academia and industry. If you are a student or ML researcher, save those dates and stay tuned for updates! 🚀
13
19
267
17,966
Samuel Simko retweeted
Apr 9
Excited for our "Trustworthy AI for Good" (AI4GOOD) Workshop at #ICML2026! As AI agents increasingly affect our lives, it is key to bridge #ResponsibleAI, social good, and governance. Let’s build solutions together!
⏰ Submission deadline: April 30, 2026 (AoE)
🎙️Confirmed speakers: @Yoshua_Bengio, Joel Z. Leibo (@jzl86), Maksym Andriushchenko (@maksym_andr), @OanaIgnatRo [More to come!]
📍July 10-11, 2026 · Seoul🇰🇷
🔗 trustworthy-ai-for-good.gith…
📝 Submit: openreview.net/group?id=ICML…
📣 Be a reviewer: forms.gle/7cXvUJCW1FdEghi6A
3
30
164
13,598
Samuel Simko retweeted
Apr 8
What if the most dangerous AI isn’t rogue - but works as intended?
A new ELLIS-affiliated paper shows aligned, policy-compliant AI can still undermine democracy at scale.
Bottom line: alignment ≠ safety. Democratic resilience must keep pace.
📄 Paper: bit.ly/4snKdLN
ALT A recent paper, co-authored by several ELLIS-affiliated researchers, shows that perfectly aligned, policy-compliant AI can undermine democratic institutions. Not by malice, but through sheer scale.
1
11
941
Samuel Simko retweeted
Apr 6
📢We will present 5 papers to #ICLR2026, #CLeaR2026, and #ACL2026:
- SocialHarmBench by @psyonp et al.
- Causal LLMs on Instrumental Variable Method by @ivakshi_s et al.
- LLM Data Contamination study by @TerryJCZhang et al.
- Mech Interp for VLM by @francescortu et al.
- DPO data selection method by Xuan & @rongwu_xu
Thanks to all our collaborators and institutional support from @MPI_IS @ELLISforEurope @UofTCompSci @VectorInst @TorontoSRI @CIFAR_News @JinesisLab @EuroSafeAI @ELLISInst_Tue @ETH_en @ETH_AI_Center @michigan_AI @UMichiganAI @UMichCSE!
Feel free to access the papers at
arxiv.org/abs/2510.04891
arxiv.org/abs/2602.07943
arxiv.org/abs/2509.00072
arxiv.org/abs/2507.13868
arxiv.org/abs/2508.04149 🎉
1
10
77
5,804
Samuel Simko retweeted
Apr 6
Our work on multilingual benchmark translation got accepted to @aclmeeting 2026 Findings! 🎉
The glorious EEU promo is coming to San Diego this summer🥺🇺🇦🇧🇬
#ACL #ACL2026 #ACL2026NLP #NLProc
Feb 26
❓Can we actually trust the quality of the existing multilingual benchmarks translated from English? Turns out many of them have some simple bugs, which hurts the evaluations - we try to fix that!
Introducing Recovered in Translation 🌍
ritranslation.insait.ai
🧵below
1
3
14
956
Mar 25
🚨 New paper: AI Poses Risks to Democratic and Social Systems.
We discuss 7 failure modes showing how Al can degrade democracy and society through power concentration, narrowing how we think, or flooding institutions faster than they can keep up.
We also proposed 7 research & governance recommendations, from simulation-based stress-testing to deliberative governance infrastructure.
Honored to work with Yoshua Bengio, Stuart Russell, Roger Grosse, Bernhard Schölkopf, Rada Mihalcea, Ashton Anderson, Audrey Tang and many others.
Full whitepaper here: zhijing-jin.com/d/2026-ai-ri…
Mar 25
AI is threatening our democratic society—by concentrating power, narrowing how we think, and flooding institutions faster than they can keep up. These risks emerge at the system level, and technical work alone won't fix them.
👉Check out our whitepaper with 25 researchers: zhijing-jin.com/d/2026-ai-ri…
💡We introduce 7 threat models and ways forward.
✍️Led by @davidguzman1120 with @DaveRBanerjee, @blin_kevin, @PepijnCobben, @gcorsi_, @x_angelohuang, @ChanglingXavier, Suvajit Majumder, @psyonp, @SimkoSamuel, @strauss_irene, and @TerryJCZhang
Advised by senior co-authors: @ashton1anderson, @Yoshua_Bengio, @MatthiasBethge, @RogerGrosse, Karoline Helbig, @david_lie, Richard Mallah, @radamihalcea, Susan Nesbitt, Susan Perry, @presnick, Stuart Russell, @mrinmayasachan, @bschoelkopf @audreyt and @ZhijingJin
Thank you to all the institutional support from @JinesisLab @EuroSafeAI @MPI_IS @CIFAR_News @iapsAI @CARMA_411 @Cambridge_Uni @UofTCompSci @VectorInst @TorontoSRI @Mila_Quebec @LawZero_ @uni_tue @michigan_AI @UMichCSE @AUParis @UNESCO @UCBerkeley @ETH_en @ETH_AI_Center @ELLISInst_Tue @ELLISforEurope @EthicsInAI #CivicAI
#AISafety #AIGovernance #Democracy #ResponsibleAI
2
9
732
Samuel Simko retweeted
Mar 22
Mech interp or representation interp?
We need to decode the causal computational graph of #LLMs—not just cataloguing representations (steering vectors etc).
Analogy: we can’t understand biology by just blood composition. We need to understand how the body works.
Same for LLMs.
4
24
165
9,892
Mar 3
🚀 We're launching EuroSafeAI, a nonprofit focused on multi-agent AI safety research, here in Zurich.
Our launch event is on 6 March at 6:30 PM at the ETH Student Project House. Expect lightning talks and drinks!
Info & Sign-up:
luma.com/hwo46ach
See you there! 🔥
1
3
11
865
Mar 3
🗓️ 6 March 2026 · 6:30 PM
📍 ETH Student Project House E floor, Clausiusstrasse 16
eurosafe.ai.toronto.edu/
1
2
99
Mar 3
Speakers:
@ZhijingJin (CIFAR AI Chair; Professor@UoT, Chief Scientist),
@x_angelohuang (Co-founder and Director), @pepijncobben (Co-founder and Director), @SimkoSamuel (Technical Staff),
@davidguzman1120 (Technical Staff).
1
80
Samuel Simko retweeted
Check out CircuitLab🚀
A Scalable Python library for training Cross-Layer Transcoders (CLTs) Visual interface & auto-interp incoming so mark our repo: github.com/circuits-research…
Collaborative effort @MPI_IS @ELLISforEurope @UofTCompSci @VectorInst @TorontoSRI @CIFAR_News @JinesisLab
4
16
2,077