Khanna to introduce AI safety bill awaiting technology safeguards

Rep. The article states that the bill includes strict liability standards and a ban on 'recursive' self-improving AI until the government creates safeguards. The article identifies the measure as an AI safety bill and states that the bill would also create a new federal agency.

According to the article, that agency would focus on AI models from OpenAI, Anthropic, Google DeepMind and xAI. The responsibilities described for the agency include safety regulations, a licensing system, frontier model audits and human-control standards. The excerpt does not state how the agency would be funded or staffed, or when it would begin operating.

The excerpt does not state when Khanna intends to introduce the bill, what penalties the strict liability standard would impose, or how the licensing system would be administered. It does not define 'recursive' self-improving AI or state how such a ban would be enforced, whether the four named companies would be required to register with the new agency, or what authority that agency would hold over them. The excerpt includes no comment from Khanna's office, from OpenAI, Anthropic, Google DeepMind or xAI, or from any existing federal agency. It does not state which congressional committee would consider the bill, whether it has co-sponsors, or when a vote might occur.