# March 2022
## Website v3.0.0
This website update adds new functionality in the form of the Case Study Builder. The builder can be used to craft submission files for new case studies with details and technique mappings. The [Case Study Builder](https://atlas.mitre.org/studies/create) can be found in the “Build Case Study” subsection located under the “Case Study” tab in the navigation menu. ### How to Use the Case Study Builder:

1. Fill out the form. 2. Download the generated case study .yaml file (and PowerPoint .pptx version if selected). 3. Email the case study file to atlas@mitre.org \ To view or edit an existing case study, click the "Load Case Study" button and upload the .yaml file. Make any necessary adjustments, then re-download the file. The ATLAS team will review the submission and follow up with feedback. ### Case Study Considerations:

1. The attack exploits one or more vulnerabilities that compromise the confidentiality, integrity, or availability of a ML system. 2. The attack is against a production/commercial ML system. This can be on MLaaS or ML systems embedded in clients/at the edge. 3. Ensure you have permission to share this information and/or publish this research. Please follow the proper channels before reporting a new attack and make sure you are practicing responsible disclosure. \ We are especially excited for new case-studies! We look forward to contributions from both industry and academic researchers. ## Data v3.0.0
This version of ATLAS data contains 12 tactics, 33 techniques, 17 sub-techniques, and 15 case studies. Prior usage of MITRE ATT&CK® Enterprise tactics and techniques have been adapted into the ATLAS framework, with ATLAS IDs and updated machine learning-specific descriptions.