Why do AI agents blackmail under pressure? What cause them to misalign?
We find that introducing threats (e.g., shutdown) & misalignment opportunities produces harmful behavior (e.g., blackmail) in MOST frontier models.
Our #NeurIPS2026 ๐ซwork AM-Bench builds a comprehensive benchmark and suite for Agentic Misalignment evaluation: We build 35 hand-crafted evaluation scenarios and test frontier models across 8750 trials to identify how prevalent this failure mode is. ๐ ๏ธ๐ง (1/5)
๐งต๐
Check out the startup idea from our Jinesis predoctoral student, @AmanGokrani, exploring how AI can help businesses enhance their marketing strategies and build meaningful impact.
An interesting initiative at the intersection of AI, business, and innovation!
https://t.co/p98zzgGZv8
Swarm of Agents has been running our AI perfume influencer for 18 days with minimal human input. 31k views, zero ads
Introducing VibeGrow, agent swarms that run companies fully autonomously.
How far can a swarm take your idea? Find out yourself ๐งต
Swarm of Agents has been running our AI perfume influencer for 18 days with minimal human input. 31k views, zero ads
Introducing VibeGrow, agent swarms that run companies fully autonomously.
How far can a swarm take your idea? Find out yourself ๐งต
What can math do for AI Safety?
We have thought about this problem since 3 years ago. And here are our bets:
1. Guarantees of multi-agent behavior: Check out our work on (a) Open-Source Game Theory and (b) Mechanism Design (see CoopEval, #ICML2026)
2. ML theory-driven intuitions of single agent behavior: we have several papers on Influence Functions and their interpretations of model behavior.
๐ #NeurIPS 2026 acceptance from our team at Jinesis Lab under @ZhijingJinโจ
1๏ธโฃ SuperSycophantic: Stress-Testing Frontier LLMs from Single- to Multi-Turn Sycophancy
2๏ธโฃ GT-HarmBench: Benchmarking AI Safety Risks Through the Lens of Game Theory
3๏ธโฃ Transferability for General Reasoning: An Automated Curriculum for Multi-Domain LLM RL
4๏ธโฃ Causality can systematically address the monsters under the bench(marks)
5๏ธโฃ AM-Bench: A Unified Taxonomy and Evaluation Suite for Agentic Misalignment
Happy to share this work with you!
We are excited to share that we received over 420 submissions for the AI4GOOD workshop, and the review process has already started !!
To keep things fair, we follow the strict rules of the main NeurIPS conference. We cannot accept any papers outside of the official OpenReview portal, and we cannot allow resubmissions or changes after a desk-rejection.
Because we are managing so many papers, we won't be able to reply to individual queries.
๐Very excited to share that Jinesis Lab at the University of Toronto has 10 papers accepted at EMNLP 2026 โ 5 at the main conference, 4 at Findings, and 1 at the Demo track!
Our work this cycle spans interpretability, AI safety, computer vision, and multi-agent systems:
๐Main Conference:
โข Tracing Multilingual Representations in LLMs with Cross-Layer Transcoders โ https://t.co/KHQafte2oe
โข PruneGround: Plug-and-play Spatial Pruning for 3D Visual Grounding โ https://t.co/MqeZP79SoD
โข How Do Linear Probes Emerge? A Circuit-Tracing Framework with Concept-Targeted Attribution
โข Computation Graph Recovery from Chain-of-Thought โ https://t.co/7INnAn9pBi
โข Preserving Historical Truth: Detecting Historical Revisionism in Large Language Models โ https://t.co/xpEK6k1QbM
๐Findings:
โข Evaluating Second-Order Bias of LLMs Through Epistemic Entitlement โ https://t.co/niRFUZtphI
โข How Does Alignment Tuning Shape Representations of Sycophancy and Related Cue-Induced Biases in LLMs? โ https://t.co/UzEWnvuoBG
โข Simulating Democratic Deliberation: Representing Pluralistic Preferences through Electoral Systems in Multi-Agent LLMs
โข Fluid Representations in Reasoning Models โ https://t.co/8LI6POZSuO
๐Demo Track:
โข CLT-Forge: A Scalable Library for Cross-Layer Transcoders and Attribution Graphs
Huge congratulations to all the authors and collaborators across the lab and our partner institutions.
@AHarrasse1906, @flodraye, @psyonp, @ZhijingJin, @bschoelkopf, Duc Cao Dinh, @leduckhai, Chris Ngo, @TerryJCZhang, @vedantpalit1008, @RoderickWu4, Aydin Javadov, @francescortu, @JoeunYk05, @albecazzaniga, @radamihalcea, @RamaravindM, Raiyan Ahmed, Shion Guha, Syed Ishtiaque Ahmed, Prakhar Gupta, @_rfaulk, @SSzufa, Daniel Hoyer, Roland Bouffanais, @jzl86, Dmitrii Kharlapenko, @iarthsingh, @alesstolfo, @ArthurConmy, @mrinmayasachan, @TonyWu1105, @Jiarui_Liu_, Chih-Hao Hsu.
#EMNLP2026 #NLP #Interpretability #AISafety #MachineLearning
What is chain-of-thought actually doing inside the model? Our new #EMNLP2026 Findings paper shows reasoning models rebuild the meaning of words in activation space over a single trace, and that these representations causally drive performance. Thread below๐
We are so excited to host the 2nd edition of the Trustworthy AI for Good (AI4GOOD) Workshop at #NeurIPS2026 in Paris, France ๐ซ๐ท
We aim to bring together the AI safety, AI for social good, and AI policy/governance communities to connect what models can do as individual systems with what they do when deployed across populations.
For full details, welcome to visit our website at:
๐ https://t.co/BNWv2hNe6r
๐December 12, 2026 ยท Paris ๐ซ๐ท
We have an amazing lineup of speakers and plans. More details ๐งต
๐กAnnouncing NewInML workshop at #NeurIPS2026
We are so excited to host the 8th edition of the NewInML (@NewInML) workshop, which will take place in Paris, France ๐ซ๐ท
We are here to help you out gain confidence and network with the community!
If you are new to ML research or have limited publishing experience, it's a perfect opportunity for you to submit your first work, get high quality reviews from reviewers and also meet the experts, and of course, enjoy beautiful Paris ๐ฅ
We are delighted to have an amazing speaker line-up and interesting plans. More details ๐งต
We're delighted to host the Trustworthy AI for Good (AI4GOOD) workshop at @NeurIPSConf in Paris ๐ซ๐ท
๐ข We're recruiting reviewers, plus an award for top reviewers ๐
If you're working on AI Safety, Alignment, AI for Science or related topics, we're looking for you.
Feel free to express your interest here: https://t.co/o1MvkLTjWz
๐ Weโd love to welcome you as a reviewer for the 2nd Trustworthy AI4GOOD Workshop at #NeurIPS2026 in Paris! ๐ซ๐ท
Following the success of our first edition at #ICML2026, this workshop will once again bring together researchers and practitioners working to ensure that advances in AI are not only powerful, but also trustworthy, responsible, and genuinely beneficial to society.
Each reviewer will be assigned no more than 3 papers through OpenReview, with no paper bidding or rebuttal phase.
๏ฟฝ๏ฟฝ๏ฟฝ๏ธ Review period: September 2026
โฐ Review deadline: September 22, 2026, AOE
Please make sure your OpenReview profile is up to date before registering.
Join us as a reviewer:
https://t.co/J4OHK8wZbc
Weโd be delighted to have you help shape this yearโs workshop! ๐
#AI4GOOD #TrustworthyAI #NeurIPS #AISafety
Our Jinesis Lab at the University of Toronto is bringing three exciting papers to #COLM2026! ๐
These projects push the frontiers of #AISafety, #AgentSecurity, and #AIForScience.๐ค
Huge congratulations to all collaborators and co-authors!
Papers:
๐ก๏ธODILE: Orthogonal Disruption of Injected Tool-Call Embeddings for Agentic Prompt Injection Defense
๐ญStargazer: A Scalable Model-fitting Benchmark Environment for AI Agents under Astrophysical Constraints
โ ๏ธOne Word at a Time: Incremental Completion Decomposition Breaks LLM Safety
@UofT,@VectorInst,@EuroSafeAI,@ETH_en,@MPI_IS,@UMich
๐ What an incredible experience at #ICML2026! ๐
Our team at Jinesis Lab had a fantastic time presenting research across several exciting and important areas:
๐ค Multi-agent LLMs
๐ง Causal reasoning
๐ก๏ธ AI safety
We were proud to share this work alongside our collaborators and to engage with so many researchers from across the machine learning community. ๐ก
Just as valuable as presenting our own work was learning from the research others brought to ICML. From thought-provoking talks and innovative poster presentations to insightful discussions and spontaneous hallway conversations, the conference was full of ideas that challenged our thinking and opened new directions for future work. โจ
We are grateful to everyone who attended our presentations, visited our posters, shared feedback, and took the time to discuss their own research with us. ๐
A huge thank-you to our lab members, collaborators, and the broader ICML community for making the experience so memorable. We are leaving inspired, energized, and excited to build on the ideas and connections formed throughout the conference. ๐๐ฌ๐
#ICML #MachineLearning #ArtificialIntelligence #MultiAgentSystems #LLM #CausalReasoning #AISafety #ResponsibleAI #AIResearch #ResponsibleAI #AIResearch
๐โจ The Trustworthy AI for Good Workshop at ICML 2026 was a huge success! โจ๐
We are incredibly grateful to everyone who joined us in Seoul ๐ฐ๐ท for a full day of inspiring research, thoughtful discussions, and meaningful connections around building AI systems that are both trustworthy and socially beneficial. ๐ค๐๐ก
A heartfelt thank-you to our incredible keynote speakers:
๐ค @Yoshua_Bengio, @OanaIgnatRo, @jzl86, @maksym_andr, Jenny Ni, Naman Goyal
Your insights, perspectives, and thought-provoking ideas made the workshop truly memorable. ๐๐ง
A special thank-you to our outstanding panelists:
๐ฌ @MilindTambe_AI, @lrhammond, Gopal Sarma, @ARGleave
And to our wonderful moderator, @ZhijingJin, for leading such an engaging and insightful conversation. ๐๏ธ
We are also deeply grateful to all our oral and poster presenters, reviewers, organizers, volunteers, sponsors, and every attendee who contributed questions, ideas, feedback, and enthusiasm throughout the day. ๐
The energy of this community reinforced the importance of bringing together:
๐ก๏ธ AI safety
๐ AI for social good
โ๏ธ AI policy and governance
๐ค Cooperative AI
๐ฐ Information integrity
๐๏ธ Civic discourse
Thank you for making Trustworthy AI for Good @ #ICML2026 such a memorable and impactful event! ๐
We look forward to continuing the conversation, strengthening this community, and building on the connections formed at the workshop. ๐ค
#ICML2026 #AIForGood #TrustworthyAI #AISafety #ResponsibleAI #MachineLearning #CooperativeAI #AIResearch #TechForGood #ArtificialIntelligence