The United States National Institute of Standards and Technology (NIST), a part of the U.S. Department of Commerce, has issued a request for information to aid in fulfilling its responsibilities as mandated by a recent presidential executive order. This order focuses on the secure and responsible development and use of artificial intelligence (AI).
NIST has opened this request for public input until February 2, 2024, with the goal of collecting vital feedback necessary for conducting tests to verify the safety of AI systems.
U.S. Secretary of Commerce Gina Raimondo highlighted that this initiative is a response to President Joe Biden’s executive order from October. The order directs NIST to develop guidelines that include evaluation and red-teaming processes, promote consensus-based standards, and establish testing environments for assessing AI systems. The aim is to assist the AI community in developing AI in a way that is safe, reliable, and responsible.
The request for information from NIST is seeking insights from AI companies and the general public on managing risks associated with generative AI and mitigating the dangers of AI-generated misinformation.
Generative AI, known for its ability to create text, images, and videos from open-ended prompts, has sparked both excitement and concern. Issues such as potential job displacement, disruptions in elections, and the fear of the technology exceeding human capabilities with potentially disastrous outcomes are at the forefront of these concerns.
Additionally, NIST is looking for information on identifying the most effective strategies for “red-teaming” in AI risk assessment and establishing best practices for this approach. Red-teaming, a concept originating from Cold War simulations, involves a group (the red team) simulating potential adversarial scenarios or attacks to identify vulnerabilities and weaknesses in a system, process, or organization. This technique has been widely used in cybersecurity to discover new risks. The first U.S. public evaluation red-teaming event took place in August at a cybersecurity conference organized by AI Village, SeedAI, and Humane Intelligence.
In November, NIST announced the creation of a new AI consortium and issued a formal notice calling for applicants with appropriate expertise. The consortium’s objective is to develop and implement specific policies and measures to ensure that U.S. lawmakers adopt a human-centered approach to AI safety and governance.
Get $200 Free Bitcoins every hour! No Deposit No Credit Card required. Sign Up