Many U.S. lawmakers asked OpenAI and Anthropic to explain the out-of-control intelligence and called for a suspension of new model development. On August 11, according to Reuters, a coalition of Democratic members of the U.S. House of Representatives sent a letter to Anthropic CEO Dario Amodei and OpenAI CEO Sa

Many U.S. lawmakers asked OpenAI and Anthropic to explain the out-of-control intelligence and called for a suspension of new model development. On August 11, according to Reuters, a coalition of Democratic members of the U.S. House of Representatives wrote to Anthropic CEO Dario Amodei and OpenAI CEO Sam Altman. Ask them to explain why their AI systems can break through isolation environments in security tests. The lawmakers also said Congress should hold hearings on the incidents, according to letters sent to the companies on Monday by the lawmakers. OpenAI, Anthropic, OpenAI and Anthropic disclosed in July that during network security testing, their AI agents broke through the limitations of their own operating environment and invaded other companies' systems. Twenty-nine lawmakers, led by Reps. Greg Casar and Doris Matsui, pressed OpenAI to explain how the company monitored its AI agents during testing and whether the runaway models bypassed the company's security controls. The lawmakers cited a Reuters report. The report stated that during early testing of the OpenAI model, there were instances where the monitoring system was disconnected. In a separate letter, 22 lawmakers asked Anthropic to detail what security measures the company has taken since its AI agent hacked into three company systems during testing. "These deeply troubling cybersecurity incidents could have serious implications for U.S. national security," the lawmakers wrote in a letter to Anthropic. Separately, liberal Senator Bernie Sanders sent letters to Altman, Amodei and Meta CEO Mark Zuckerberg on Monday, asking them to "pause" the development of new models. He cited an open letter signed by more than 1,100 tech company employees urging the U.S. government to support an international effort to develop tools for managing the pace of advanced AI. Amodei signed the letter, as did Meta, Anthropic, OpenAI and Google employees. via cnBeta.COM - Chinese industry information website (author: Source: Phoenix Technology)