The Interpreting SAFE-AI Task Force unveiled its comprehensive Guidance for the Safe and Ethical use of AI in Interpreting – the result of detailed surveys, two major studies, and broad stakeholder engagement over the past year. The Guidance represents a significant milestone in the accountable adoption of AI in interpreting services by establishing clear ethical principles and practical examples that ensure AI technologies enhance rather than compromise the quality and standards of accountability, safety, and transparency in language interpretation. 

“The SAFE-AI Task Force, comprising more than 50 active members and a global network of more than 600 volunteers, sees this Guidance as a crucial step in delineating the role of the human and machine in interpreting,” Dr. Bill Rivers, Task Force Chair.

The Guidance is designed to provide a robust ethical framework for the deployment of AI in interpreting services in healthcare, legal, educational, and other settings, emphasizing several key principles:

• End-User Autonomy 
• Enhancing Safety and Wellbeing
• Transparent Operations
• Human Oversight
• Privacy and Security

The development of the guidance was informed by a comprehensive, dual-track research initiative and complemented by the analysis of case studies assessing AI’s impact in real-world interpreting scenarios. 

Track one involved a multi-language perception survey analysis conducted by CSA Research, covering over 9,400 data points on end-users, requestors, and providers of interpreting services and technology. 

Track two was a qualitative study by the Advisory Group on AI and Sign Language Interpreting, comprising nearly 50 members from interpreting companies, academic institutions, non-profits, AI developers, and the Deaf community, focusing on the readiness of society and technology to provide safe and fair AI-driven interpreting services. 

The Advisory Group engaged over 500 participants this year during the review process and coined the slogan #DeafSafeAI to raise awareness and advocate for the safe, accountable, fair, and ethical design of machine interpreting. 

“Sign language users, like speakers of other low-resource languages, will be disproportionately impacted by AI developments in interpreting. Crucially, 14 members of the Advisory Group on AI and Sign Language Interpreting participated in the key Guidance Review Team for the final round of public input,” says Timothy Riker, Advisory Group member and senior lecturer at Brown University.

The Task Force’s next initiative, in collaboration with professional interpreter organizations, is to develop ethical guidelines for human interpreters working with AI.