IEEE SaTML 2027 — Conference on Secure and Trustworthy Machine Learning

IEEE SaTML 2027, the Conference on Secure and Trustworthy Machine Learning, is scheduled for 8-10 May 2027 in Reykjavik, Iceland — the official site currently states the timing as early May 2027, with exact dates confirmed through the conference deadline listings. Organised by the IEEE Computer Society Technical Committee on Security and Privacy, SaTML exists because ML security had been split between machine-learning venues that treat security as an afterthought and security venues that treat ML as a black box. Its call for papers covers attack and defence mechanisms, secure and safe implementation of ML systems, algorithm and system verification, privacy protections, forensic analysis, fairness and interpretability, and trustworthy data practices. It accepts research papers, systematization-of-knowledge papers and position papers, and runs a competitions programme and workshops alongside the main track.

• Paper submission deadlineSep 29, 2026
• Early rejections announcedNov 4, 2026
• Revision phase opensNov 25, 2026
• Revision phase closesDec 9, 2026
• Final decisionsDec 16, 2026
• Revised papers dueJan 21, 2027

Date

May 8–10, 2027

Location

Reykjavik, Iceland

Format

In-Person

Event Type

Conference

Status

Upcoming

Pricing

Paid

Organizer

IEEE Computer Society Technical Committee on Security and Privacy

Added

Aug 16, 2026

Updated

Aug 16, 2026

• ✓AI security leads
• ✓Heads of responsible AI and model risk
• ✓ML engineers deploying models in regulated settings
• ✓Security architects
• ✓Privacy engineers
• ✓Model risk and assurance managers

This is the shortest path to an evidence-based answer on whether your model deployments are defensible. A model-risk or AI-security lead leaves with the current state of extraction, poisoning, jailbreak and membership-inference attacks against production-shaped systems, the verification techniques that can be cited in an assurance argument, and the competition artefacts to run those attacks against your own models. For anyone writing an internal AI security standard, this is where its technical claims should come from rather than a vendor whitepaper.

• ✓Research paper, systematization-of-knowledge and position paper submission types
• ✓CFP area: ML attack and defence mechanisms, and adversarial robustness
• ✓CFP area: privacy protections and forensic analysis of ML systems
• ✓CFP area: algorithm and system verification for machine learning
• ✓Competitions programme, proposed and run alongside the main track
• ✓Co-located workshops proposed through the same call

• ★Purpose-built for trustworthy ML rather than an AI track inside a general security conference or a security workshop inside an ML conference
• ★Runs a revision phase (25 November - 9 December 2026) between early rejection and final decision, so borderline work is repaired rather than resubmitted a year later
• ★Hosts a competitions programme, which produces reproducible attack and defence artefacts rather than paper-only claims
• ★Position papers are a first-class submission type, which is where the arguments about evaluation standards and disclosure norms actually get made

Registration rates for SaTML 2027 are not yet published; the site currently carries the call for papers and the deadline schedule only.

When is IEEE SaTML 2027?

IEEE SaTML 2027 takes place May 8–10, 2027 in Reykjavik, Iceland.

Where is IEEE SaTML 2027 held?

IEEE SaTML 2027 is held in Reykjavik, Iceland. The event format is in-person.

What is the IEEE SaTML 2027 submission deadline?

The next IEEE SaTML 2027 deadline is paper submission deadline on Sep 29, 2026. Remaining dates: Early rejections announced — Nov 4, 2026; Revision phase opens — Nov 25, 2026; Revision phase closes — Dec 9, 2026; Final decisions — Dec 16, 2026; Revised papers due — Jan 21, 2027.

How much does IEEE SaTML 2027 cost?

IEEE SaTML 2027 is paid. Registration rates for SaTML 2027 are not yet published; the site currently carries the call for papers and the deadline schedule only.

Who should attend IEEE SaTML 2027?

IEEE SaTML 2027 is aimed at AI security leads, Heads of responsible AI and model risk, ML engineers deploying models in regulated settings, Security architects, Privacy engineers, Model risk and assurance managers.

What topics does IEEE SaTML 2027 cover?

IEEE SaTML 2027 covers Machine learning attacks and defences, Secure and safe ML implementation, Verification of ML algorithms and systems, Privacy in machine learning, Forensic analysis of ML systems, Fairness and interpretability, Trustworthy data practices, Adversarial robustness.

This page was written from 3 sources, 1 on domains other than satml.org.

1. 1.satml.org — satml.orgvendor
2. 2.satml.org — participate cfpvendor
3. 3.mldeadlines.com — satml 2027