Computer use models are attracting significant interest in academia and industry due to their ability to perform complex tasks in non-deterministic environments.

However, they are far from being ready for unattended deployment, as evidenced by their performance on the OSWorld benchmark where they achieve only a small fraction of human performance.

The rapid evolution of these agents raises important questions regarding their accuracy, safe deployment, and potential impact on the future of work. 

If you can’t make it to Vancouver, you can register for a virtual pass allowing you to view the ICLM live stream using this link.