Post-AI Formal Methods
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
All are welcome to attend the workshop - as speakers, or listeners.
No paper is required for attendance.
Call for Submissions
Workshop on Post-AI Formal Methods
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 Instructions
Please select the appropriate category when submitting your work:
- AAAI Papers (regardless of acceptance status)
- Authors should submit the AAAI paper title and abstract, together with two separate PDF files:
- Summary file (1 page): a short (1-2 paragraphs) explanation of how the work relates to the P-AI-FM workshop topics.
- Review file: the Phase I notification (if rejected, include the reviews; if accepted, the notification alone is sufficient).
- Accepted submissions will be considered for oral presentation slots (10 minutes each).
- At least one author of each accepted submission must register for the workshop.
- Short / Position Papers (2-4 pages)
- Upload a single PDF file containing your short or position paper (vision statement, preliminary results, or open problems).
- Accepted short papers will be listed on the workshop website, with a link to an open-access version (e.g., arXiv).
- Short papers will not be allocated oral presentation slots. Depending on the number of submissions, authors may be invited to give brief "lightning talks" (2-5 minutes, informal).
- At least one author of each accepted short paper must register for the workshop.
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).
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
Workshop Date: January 26 / 27, 2026 (at AAAI-26, Singapore)
Invited Speakers
Established researchers on the intersection of Formal Methods and Artificial intelligence.
The invited speakers will include leading researchers in formal methods and AI. The full list of confirmed speakers will be announced soon.
A professor at the School of Computing at the National University of Singapore. His research focus on formal methods, safety and security systems, probabilistic reasoning, and trusted machine learning. He co-founded the PAT verification system, “Silas: Trusted Machine Learning” and the Dependable Intelligence company.
A full professor at King's College London. Her research focus on formal verification, causality, and explainable AI, with applications to hardware and software systems. She pioneered the use of causality in software engineering, leading to the first industrial applications at IBM and Intel.
TBA
TBA
Organizing Committee
Andoni Rodríguez
Ph.D. student, IMDEA Software Institute.
Expertise: Formal methods in AI, logic-based verification, and dynamic systems.
Noa Izsak
Ph.D. student, Ben-Gurion University & CISPA Helmholtz Center for Information Security.
Expertise: Learning and grammatical-inference, verification of parameterized 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
Dr. -Ing. Swen Jacobs
Professor at CISPA Helmholtz Center for Information Security
Prof. César Sánchez
Full Professor at the IMDEA Software Institute in Madrid
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
Daniel K.
"I honestly received the best service I could have asked for. They were super professional to deal with; they met all my needs and helped me a lot."
Ron H.
"The entire experience was so easy and friendly, and the price was very reasonable. I don't know what I would have done without them."
James F.
"I can't say enough good things about the quality of this company. Highly, highly recommend using them!"
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–$2000):
- All Supporter benefits, plus,
- Prominent logo placement and acknowledgment in workshop communications.
- Lead Sponsor ($2000+):
- 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
