Some Generative AI Company Employees Pen Letter Wanting 'Right to Warn' About Risks
What are the potential consequences mentioned for employees who break non-disclosure agreements according to the recent internal shuffle at OpenAI?
According to the recent internal shuffle at OpenAI, employees who break non-disclosure agreements may forfeit their rights to their vested equity in the company. This could have significant financial implications for the employees, as the value of their vested equity could far outweigh their salaries.
What specific protections are the signatories of the 'Right to Warn' letter advocating for in relation to whistleblowers within AI companies?
The signatories of the 'Right to Warn' letter are advocating for several protections for whistleblowers within AI companies3. These protections include:
- Not enforcing agreements that prevent criticism of the company for risk-related concerns.
- Facilitating an anonymous process for employees to raise concerns about AI risks.
- Supporting a culture of open criticism regarding AI risks, while still respecting trade secrets.
- Not retaliating against employees who publicly share risk-related confidential information after other processes have failed.
These demands aim to promote transparency, open dialogue, and a culture of open criticism within major generative AI companies, allowing employees to voice their concerns about potential risks associated with AI without fear of retaliation.
How does the 'Right to Warn' letter suggest AI companies should handle internal criticisms and concerns about risks?
The 'Right to Warn' letter suggests that AI companies should handle internal criticisms and concerns about risks by implementing four key principles:
- Not enforcing agreements that prohibit criticism of the company for risk-related concerns1.
- Facilitating an anonymous process for employees to raise their concerns to the company's board, regulators, or independent organizations.
- Supporting a culture of open criticism in regards to risk, while allowing for the protection of trade secrets.
- Not retaliating against former and current employees who share "risk-related confidential information after other processes have failed."
The letter emphasizes the importance of transparency, oversight, and protection for employees who speak out about potential risks. It also highlights the need for a verifiable process for employees to provide anonymous feedback on their activities, as ordinary whistleblower protections are insufficient due to their focus on illegal activity, whereas many of the risks associated with AI are not yet regulated.