OpenAI says it is ‘committed’ to security in letter to US lawmakers
In response to questions from U.S. lawmakers, OpenAI said the company is committed to ensuring its powerful AI tools do not cause harm and that employees have ways to raise concerns about safety practices.
The startup sought to reassure lawmakers about its commitment to security after five senators, including Sen. Brian Schatz, a Democrat from Hawaii, raised questions about OpenAI’s policies in a letter to CEO Sam Altman.
“Our mission is to ensure that artificial intelligence benefits all of humanity, and we are committed to implementing strict safety protocols at every stage of our process,” Chief Strategy Officer Jason Kwon said in a letter to lawmakers on Wednesday.
OpenAI specifically said that it will continue to honor its pledge to allocate 20 percent of its computing resources to security-related research for several years. The company also promised in its letter that it will not enforce nondisparagement agreements for current and former employees, except in specific cases of a mutual nondisparagement agreement. OpenAI’s previous restrictions on employees who left the company have come under scrutiny as unusually restrictive. OpenAI has since said it has changed its policy.
Altman later explained the strategy on social media.
“Our team has worked with the US AI Safety Institute on an agreement that would provide early access to our next baseline model so we can work together to advance the science of AI evaluations,” he wrote on X.
A few quick updates on security at Openai:
As we said last July, we aim to dedicate a minimum of 20% of computing resources to security efforts across the company.
Our team has worked with the US AI Safety Institute on an agreement that would provide…
— Sam Altman (@sama) August 1, 2024
Kwon also referenced in his letter the recently formed Safety and Security Committee, which is currently reviewing OpenAI’s processes and policies.
In recent months, OpenAI has faced a series of controversies surrounding its commitment to security and the ability of employees to speak out on the topic. Several key members of its security-related teams, including former co-founder and chief scientist Ilya Sutskever, have stepped down, along with another leader of the company’s team dedicated to assessing long-term security risks, Jan Leike, who publicly expressed concerns that the company was prioritizing product development over security.
© 2024 Bloomberg LP
(This story has not been edited by NDTV staff and is auto-generated from a syndicated feed.)
