A group of former and current OpenAI employees penned an open letter detailing their concern about the serious risk of the rapid advancement of AI.
The letter also claimed that there is a lack of oversight despite the increasing progression of the technology.
(Photo : Pau Barrena/AFP via Getty Images)
Read Also: OpenAI Designates Safety Committee Focused on Training Latest AI Models
AI Employees Reveal Serious Risk on AI
In an open letter, the group shared that AI companies are aware and have acknowledged the risks of developing AI. In addition, government agencies and other AI experts have agreed that there are risks.
Some of the risks listed are manipulation, misinformation, and loss of autonomy to AI systems. The letter also suggests that such serious risks can potentially result in human extinction.
While AI companies have large data of non-public information about their AI systems' capabilities and limitations, they often choose to not disclose it to the public.
"They currently have only weak obligations to share some of this information with governments, and none with civil society. We do not think they can all be relied upon to share it voluntarily," the group wrote.
AI Employees Seek Principle Changes Within Companies
According to the letter, most employees are scared to publish any form of retaliation as whistleblower protections are insufficient. The group also called out the government for the lack of oversight of high-profile corporations.
The employees published some principles that they wish to witness in advanced AI companies. For instance, companies must not enforce any policy that prohibits employees from voicing criticism.
In addition, a verifiably anonymous process must be done for both current and former employees who will raise risk-related issues to the board, regulators, and independent organizations. The company should not retaliate against whistleblower employees.
AI companies are now called upon to support a culture of open criticism from their employees. The group also seeks prioritization on informing the public about any risk.
Related Article: AI Engineers Experience Burnout Amid Fast-Paced Competition