ALBAWABA - According to a report released on Thursday by the U.S. Artificial Intelligence Safety Institute, artificial intelligence behemoths OpenAI and Anthropic have inked agreements and established a new tab with the US government for the purpose of researching, testing, and evaluating their new AI models.
According to a statement from the National institution of Standards and Technology (NIST), which is situated under the Department of Commerce, it will have access to all significant new models from each of the companies both before and after they are made public.
The institution was founded in October 2023, after the issuance of the first-ever presidential order on artificial intelligence by the Biden-Harris administration, according to CNBC. The order mandated the creation of new safety evaluations, guidelines for fairness and civil rights, and study the effects of AI on the job market.
The agreements not only provide early access to models, but they also set the path for joint study into how to assess models and their safety, as well as techniques for limiting risk, according to Axios.
Elizabeth Kelly, U.S. AI Safety Institute director said in a statement “Safety is essential to fueling breakthrough technological innovation,” adding that “These agreements are just the start, but they are an important milestone as we work to help responsibly steward the future of AI.”
Anthropic co-founder and head of policy Jack Clark commented that “Safe, trustworthy AI is crucial for the technology's positive impact,” saying “Our collaboration with the U.S. AI Safety Institute leverages their wide expertise to rigorously test our models before widespread deployment.”
