News
- [7/11/2026] Workshop Day is Here — Looking forward to seeing you in Hall B2! Check out our post on X and our posts on LinkedIn (1, 2, 3, 4, 5, 6, 7, 8).
- [7/9/2026] Schedule Posted — The final workshop schedule is now available, featuring invited and contributed talks, poster sessions, and a panel discussion. See our posts on X and LinkedIn.
- [5/8/2026] Submission Deadline Today — Today is the paper submission deadline (May 8, 2026 AoE)! Authors are welcome to use the ICML, NeurIPS, or CoLM LaTeX templates, or templates from other top venues like ACL and CVPR.
- [5/4/2026] Final Call for Papers — The final call for papers is posted! Check out our X and LinkedIn posts.
- [4/25/2026] Call for Reviewers — The call for reviewers is posted! Check out our X and LinkedIn posts. We invite researchers to serve as reviewers for the workshop. Prior reviewing experience is helpful but not required; familiarity with workshop topics is expected. Please submit your application by April 30th. We plan to present Best Reviewer Awards!
- [4/10/2026] Call for Papers — The call for papers is posted! Check out our X and LinkedIn posts. Also, the submission portal is now open!
- [4/1/2026] Workshop Accepted — The workshop has been accepted by ICML! See you in Seoul!
About
AI agents are being rapidly deployed in the real world — from OpenAI's Operator to open-source tools like OpenClaw — yet safety & security research still lags behind. As these systems reason, act, and adapt in open-ended environments, they introduce profound and emerging challenges around safety, security, and trustworthiness that the research community is only beginning to face.
This workshop brings together researchers and practitioners across academia and industry to chart the next steps toward reliable agents that can operate responsibly in the wild. We build on the momentum of our first workshop at ICLR 2026, which drew 235 submissions and over 800 anticipated attendees.
Call for Papers
The Second Workshop on Agents in the Wild at ICML 2026 invites submissions from researchers and practitioners exploring how intelligent agents can reason, act, and adapt safely and securely in open-ended real-world environments.
As agentic AI systems grow more capable, their deployment in dynamic, unpredictable settings introduces challenges in safety, security, and general trustworthiness. This workshop aims to spark discussion across academia and industry on methods, benchmarks, and frameworks for building reliable and trustworthy agents that can operate responsibly "in the wild."
Scope
We welcome contributions on a wide range of topics related to AI agents, including but not limited to:
- Agentic safety and alignment
- Agent security, privacy, and robustness
- Agentic hallucination and factuality
- Agentic interpretability and transparency
- Agentic fairness and bias
- Evaluating and benchmarking agents
- Multimodal and computer-use agents
- Multi-agent coordination and long-horizon safety
- Post-training and adapting agents
- Agent systems and infrastructure
- Interdisciplinary agentic considerations
- Ethics, society, and governing of agents
Important Dates
- Paper Submission Open: April 10, 2026 AoE
- Paper Submission Deadline:
May 1, 2026 AoEMay 8, 2026 AoE - Paper Notification Deadline:
May 15, 2026 AoEMay 22, 2026 AoE - Camera-ready Version Deadline: June 30, 2026 AoE
Submission Guidelines
Format: This workshop offers two seperate submission tracks:
- Regular Papers Track: The workshop welcomes submissions of research and position papers (9 pages). References and supplementary materials will not count against these limits.
- Short Papers Track: We encourage submission of short papers (4 pages) to make the workshop more accessible to researchers outside the ML conference publication circuit. These submissions can present implementations of unpublished ideas, modest theoretical results, follow-up experiments, or fresh perspectives on existing work. References and supplementary materials do not count against the 4-page limit.
Submission site: Submit papers through the Workshop Submission Portal on OpenReview. Please be sure to create an OpenReview profile at least two weeks in advance of the paper submission deadline.
Style file: You are welcome to use the ICML, NeurIPS, or CoLM LaTeX templates, or templates from other top venues like ACL and CVPR. For convenience, we provide a modified ICML template that refers to our workshop: download here. Submissions that exceed the page limit may be desk-rejected.
Dual-submission policy: The workshop will adopt a non-archival policy, welcoming ongoing and unpublished work, as well as papers under review or recently accepted at other venues (provided they do not breach dual-submission or anonymity policies of the other venue). Workshop submissions can be subsequently or concurrently submitted to other venues.
Visibility: Accepted papers will be made public, but rejected submissions and reviews will not.
Double-blind reviewing: Submissions must be fully anonymized. This policy applies to any supplementary or linked material as well, including code. Any papers found to be in violation of this policy may be desk-rejected.
LLM usage policy: AI-generated papers are not allowed. AI assistance is permitted, but submissions must be primarily human-authored, reflecting original thought and analysis.
Contact: For any questions, please contact us at agentwild-workshop-icml2026@googlegroups.com.
Schedule
All times are local.
| 8:00–8:10 | Opening Remarks — Dawn Song |
| 8:10–8:40 | Invited Talk 1 — Vincent Sunn Chen Art & Science of Benchmarking Agents |
| 8:40–9:10 | Invited Talk 2 — Yoshua Bengio Non-Agentic Honest Predictors as a Building Block for Safe Agents |
| 9:10–9:40 | Invited Talk 3 — Jiantao Jiao PivotRL: High Accuracy Agentic Post-Training at Low Compute Cost |
| 9:40–10:10 | Invited Talk 4 — Ying Liu Towards Autonomous and Self-Improving Agents |
| 10:10–10:40 | Contributed Talks
|
| 10:40–12:00 | Poster Session 1 Hall A, Posters 100–126, 200–217, 300–309 |
| 12:00–13:00 | Lunch Break |
| 13:00–13:50 | Panel Discussion The Future of Safe and Secure Agents Moderator: Yizhou Sun Panelists: Frederic Sala, Madhu Sehwag, Huan Sun, Eric Wallace, Chi Wang, Fei Zhang |
| 13:50–14:10 | Invited Talk 5 — Chi Wang My AI Stand: Realtime by Day, Rewrite Itself by Night |
| 14:10–14:30 | Invited Talk 6 — Wei Wang (Re)visiting Foundation Models for Science and Beyond |
| 14:30–15:00 | Invited Talk 7 — Eric Wallace Training Robust Agents |
| 15:00–15:30 | Invited Talk 8 — Michael Bendersky Thinking Fast & Slow: How Databricks Built High-Speed and Deep Research Agents |
| 15:30–15:40 | Awards & Closing — Chenguang Wang |
| 15:40–17:00 | Poster Session 2 Hall A, Posters 100–126, 200–217, 300–309 |
Invited Speakers
Panel
Workshop Organizers
Sponsors
