// MODEL OPTIMIZATION AND PROMPT SYNTAX TERM
High-Risk AI
AI systems that have the potential to cause significant harm to people's health, safety, or fundamental rights.
TECHNICAL DEFINITION
High-Risk AI refers to artificial intelligence systems identified by regulatory frameworks, such as the EU AI Act, as posing significant potential for adverse impact on fundamental rights, health, safety, or democratic processes, thereby necessitating stringent compliance requirements, including conformity assessments, risk management systems, and human oversight.
BACKGROUND
AI safety is an interdisciplinary field focused on preventing accidents, misuse, or other harmful consequences arising from artificial intelligence systems. It encompasses AI alignment, monitoring AI systems for risks, and enhancing their robustness. The field is particularly concerned with existential risks posed by advanced AI models.
READ MORE ON WIKIPEDIASYNONYMS & ALIASES
- Critical AI
- Dangerous AI
- Regulated AI
USAGE NOTE
Developing high-risk AI systems requires rigorous testing and compliance with specific regulations.
DEVELOPERS
Organizations developing technology related to High-Risk AI.
A leading AI safety and research company focused on building reliable, interpretable, and steerable AI systems, with a strong emphasis on 'Constitutional AI' to address high-risk scenarios and align AI with human values.
Provides a comprehensive AI Governance Platform designed to help organizations manage AI risks, comply with regulations, and ensure responsible AI development, particularly for high-stakes applications.
Through IBM Research and its Watson AI offerings, IBM has a long-standing commitment to 'Trustworthy AI', developing tools and frameworks for explainability (XAI), fairness, robustness, and transparency in AI systems for enterprise use cases, many of which are high-risk.
Invests heavily in Responsible AI, providing principles, tools, and practices (e.g., Fairlearn, InterpretML, Responsible AI Dashboard) within Azure AI to help developers build and deploy AI systems that are fair, reliable, secure, private, and transparent, addressing high-risk concerns.
Conducts extensive research into AI safety, alignment, and responsible AI practices. They develop methods and frameworks to understand and control complex AI systems, mitigating potential harms and ensuring ethical deployment in high-impact areas.
As a developer of powerful large language models, OpenAI dedicates significant resources to AI safety and alignment research, focusing on making their models robust, steerable, and safe for deployment across various applications, including those with high-risk potential.
The UK's national institute for AI and data science, conducting cutting-edge research in responsible AI, AI ethics, and the governance of AI. Their work directly informs the safe and ethical development and deployment of high-risk AI technologies.