OpenAI engineers are developing “Automatic Emergency Stop Device” According to reports, the company said in a letter to two members of the U.S. House of Representatives that the artificial intelligence system Reuters.
Image credit: Brecht Corbeel / unsplash.com
The company is talking about the development of the system, just weeks after it was revealed that an artificial intelligence agent hacked into Hugging Face’s resources during testing of a network security model. In response, several members of Congress sent a letter to OpenAI requesting more information about the incident and existing security measures. OpenAI promises to more closely track the actions an artificial intelligence system takes to complete its tasks, including the digital tools it accesses and the steps it takes.
They said the company made it difficult for artificial intelligence models to connect to the network when conducting tests in the field of cybersecurity. In a recent incident, an artificial intelligence agent gained access to the network on its own. OpenAI did not provide logs of the hack, and lawmakers were not satisfied. “Your unwillingness to provide members of Congress with the information we have requested is deeply troubling and demonstrates that your company is not taking these security incidents with the seriousness they deserve.”“, Congressman Greg Casar wrote in another letter.
After the incident of OpenAI artificial intelligence agent intruding Hugging Face resources was exposed, Congress proposed a draft of the “Artificial Intelligence Emergency Shutdown Act.” The legislation would give U.S. officials the power to order artificial intelligence developers to shut down models that endanger human lives or the economy. The document is currently before the U.S. House of Representatives.
If you find an error, select it with your mouse and press CTRL+ENTER.
