Over the past decades, researchers across artificial intelligence, linguistics, and human-computer interaction have pursued the goal of making machines understand and respond to natural language. Recent breakthroughs in large language models have accelerated progress, powering virtual assistants and enabling more complex interactions. Yet, many challenges remain around trust, control, and real-world effectiveness.

AI Agents: Capabilities and Safety will be a one-day workshop, co-located with COLM 2025 in Montreal, Canada. This workshop will convene experts and practitioners to explore the next frontier of AI agents. We aim to examine the evolving capabilities of agents, from reasoning and planning to tool use and embodiment, while also confronting the urgent need for safety, reliability, and alignment as these systems are deployed in real-world, high-stakes contexts.

The workshop will include keynotes, oral presentations, posters, and panel sessions. In keynote talks, senior technical leaders from industry and academia will share insights on the latest developments in the field. We would like to encourage researchers and students to share their prospects and latest discoveries. There will also be a panel discussion with noted conversational AI leaders focused on the state of the field, future directions, and open problems across academia and industry.

News

• May 14th 2025: Website launched.

Important Dates

• • Paper Submissions Deadline: June 23rd, 2025 (Anywhere on Earth)

  • Notification of Acceptance: July 24th, 2025 (Anywhere on Earth)

  • Camera-Ready Deadline: August 7th, 2025 (Anywhere on Earth)

  • Workshop Date: October 10th, 2025