Post-AI Formal Methods
The Workshop on Post-AI Formal Methods at AAAI-26
About
Exploring the Future of Post-AI Formal Methods
As AI systems increasingly emerge through automated training, optimization, and generation, rather than direct human programming or design, new challenges arise to ensure their correctness, reliability, and trustworthiness.
This workshop brings together researchers working at the intersection of formal methods and machine learning. It aims to foster dialogue around the new challenges and opportunities posed by systems we didn't fully build - but which we must still understand, trust, and reason about.
Topics of interest
(including but are not limited to)
- Formal verification of AI systems / hybrid pipelines
- Ensuring Correctness of AI-Generated Models
- Verification of Concurrent / Multi-agent / Distributed Systems
- Automata learning and inference for black-box systems
- Counterexample-guided learning or synthesis
- Symbolic representations for explainability and safety
- Neuro-symbolic methods - their correctness foundations
- Runtime monitoring and formal auditing of AI behavior
- Applications of formal methods in real-world AI systems (e.g., robotics, NLP, healthcare)
Participation/Attendance
Attendance is open to all registered participants of AAAI-26.
To present at the workshop, authors must have a submission accepted in one of the categories below, and at least one author of each accepted submission must register for the workshop.
Call for Submissions
Workshop on Post-AI Formal Methods (P-AI-FM)
We invite submissions for contributed talks to be presented at the Post-AI Formal Methods workshop. The goal of this workshop is to foster interaction between the AI and formal methods communities, with contributions spanning verification, synthesis, automata learning, explainability, and neuro-symbolic approaches.
We especially encourage both young and experienced researchers to participate, present their work, share their perspectives, and connect with colleagues in the community - particularly those exploring the intersection of symbolic reasoning, formal modeling, and data-driven AI.
Submission Tracks
Track 1: AAAI Papers and Recently Published Work:
This track welcomes submissions in the following categories:
- Papers submitted to the AAAI-26 main track (regardless of acceptance status).
- Already published papers, or papers accepted but not yet formally published, in conferences such as CAV, TACAS, FM, AAAI, IJCAI, NeurIPS, ICML, ICLR, AAMAS, POPL or other related venues (2024 and later).
Submission Requirements:
- Paper title and abstract.
- Summary file (up to 1 page): a short explanation (1–2 paragraphs) of how the work relates to the P-AI-FM workshop topics.
- Review/Proof file:
- For AAAI-26 submissions: the Phase I notification (if rejected, include the reviews; if accepted, the notification alone is sufficient).
- For already published or accepted work: proof of acceptance (e.g., bibliographic citation, acceptance notice, or official link).
Accepted submissions will be considered for
oral presentations
(10-15 minutes).
At least one author of each accepted submission must register for the workshop.
Track 2: Exploratory and Published Papers
This track is intended to encourage a wide range of contributions, including:
- Extended abstracts.
- Already published work (that aligns with the P-AI-FM topics).
- Preliminary results.
- Vision statements, early ideas, or open problems.
Submission requirements:
- A single PDF file containing the contribution. There is no specific length restriction; contributions of up to 10 pages are recommended (excluding appendix and bibliography).
- A Summary file (up to 1 page): a short explanation (1–2 paragraphs) of how the work relates to the P-AI-FM workshop topics.
Accepted contributions will be listed on the workshop website, with a link to an open-access version (e.g., arXiv). Depending on the number of submissions, authors may be invited to give ''lightning talks'' (5-10 minutes).
At least one author of each accepted contribution must register for the workshop.
Clarification on Presentation Policy:
- In accordance with AAAI-26 policy, all in-person workshop presentations must be delivered onsite in Singapore. Remote or recorded presentations cannot be shown during the live sessions.
- However, to increase the visibility of contributions from authors unable to attend in person, we warmly invite such authors to optionally submit a 5–10 minute pre-recorded video (MP4 format). These videos will be uploaded to the official P-AI-FM workshop website and made available to the broader research community.
- This option provides an additional channel of dissemination, but please note:
- The recording will not be played during the workshop itself.
- Registration is still required for all accepted contributions, regardless of in-person attendance.
General Notes
- All submissions must be in PDF format.
- Submissions will be checked for relevance to the workshop themes.
- Deadlines are firm (Submission: October 22, 2025 AoE; Notification: November 5, 2025 AoE).
Frequently Asked Questions (FAQ)
Important Dates
Workshop on Post-AI FM
Please be aware that submission deadlines are firm and set in accordance with AAAI's overall schedule. Unfortunately, this means that we cannot offer flexibility with late submissions. We therefore encourage all authors to prepare their materials in advance and submit on time to ensure consideration
Submission Deadline:
October 22, 2025 (AoE)
Notification:
November 5, 2025 (AoE)
Workshop Date:
January 26, 2026 (at AAAI-26, Singapore)
Invited Speakers
Established researchers on the intersection of Formal Methods and Artificial intelligence.
Organizing Committee
Andoni Rodríguez
Ph.D. student, IMDEA Software Institute and currently at AWS intership. Expertise: Formal Methods in AI, Logic based Verification, and Dynamic Systems.
Noa Izsak
Doctoral Researcher at CISPA Helmholtz Center for Information Security. Expertise: Automata learning, Formal Methods and Verification, Reasoning about Distributed & Concurrent Systems
Djordje Zikelic
Asst. Prof., School of Computing and Information Systems, Singapore Management University (SMU). Expertise: Formal Methods for Probabilistic Models and Programs, Trustworthy AI, Safe Autonomy.
Steering Committee
Prof. Moshe Y. Vardi
Professor at Rice University and Fellow at the Baker and Ken Kennedy Institutes.
Ass. Prof. Bettina Könighofer
Assistant Professor at Graz University of Technology
Prof. César Sánchez
Full Professor at the IMDEA Software Institute in Madrid
Evaluation Committee
Evaluation Committee Chairs
- Noa Izsak (CISPA)
- Andoni Rodríguez-Barcenilla (IMDEA)
- Đorđe Žikelic (SMU)
Evaluation Committee Members
- S Akshay (IIT Bombay)
- Rayna Dimitrova (CISPA)
- Guy Amir (Cornell University)
- Filip Cano Córdoba (ISTA)
- Emily Yu (Leiden University)
- Elina Sudit (BGU)
- Daniel Jurjo
(IMDEA)
- Omri Isac
(HUJI)
Tentative Workshop Schedule
The workshop will feature a mix of invited talks, contributed paper presentations, and a panel discussion.
The exact schedule (including session times and speaker line-up) will be finalized after the review process, based on the number and type of accepted contributions.
A detailed program will be announced closer to the conference date.
Customer reviews
Sponsors
We are grateful for the generosity of the donors who made the workshop possible.
Special thanks to our Lead Sponsors
The CISPA Helmholtz Center for Information Security is a national Big Science institution within the Helmholtz Association, located in Saarbruecken, Germany. It explores information security in all its facets in order to comprehensively and holistically address the pressing major challenges of cybersecurity and trustworthy artificial intelligence that our society faces in the digital age. CISPA holds a global leadership position in the field of cybersecurity, and is internationally recognized as a training ground for the next generation of cybersecurity experts and scientific leaders in the field.
Sponsorship Opportunities
Support the Post-AI Formal Methods Workshop, AAAI-26
The Post-AI Formal Methods (P-AI-FM) workshop at AAAI-26 invites industry and academic partners to support this unique community. Formal methods for AI, while still a relatively small research area, are crucial for the future of trustworthy and reliable AI.
By sponsoring the workshop, you will help us bring invited speakers, support broad participation, and increase recognition of this important field under the official AAAI umbrella.
We offer three levels of sponsorship:
- Supporter (up to $500):
- Logo on the workshop website, verbal acknowledgment.
- Sponsor ($500–$2500):
- All Supporter benefits, plus,
- Prominent logo placement and acknowledgment in workshop communications.
- Lead Sponsor ($2500+):
- All Sponsor benefits, plus,
- Special mention during the workshop opening and closing, and
- Recognition in the official workshop report to AI Magazine.
For sponsorship inquiries, please contact us at: postaifm26@easychair.org















