Ex-OpenAI and Google employees demanded that staff be given the right to talk about the risks of AI
Date: 2024-06-06 Author: Oliver Abernathy Categories: IN WORLD
A recent open letter signed by former and current artificial intelligence (AI) experts from companies such as OpenAI and Google DeepMind raises an important question about employees' freedom to express concerns about the risks of AI. The appeal calls on the industry to allow its employees to openly discuss potential threats associated with the development and use of AI.
A Call for Openness
Experts are demanding that AI companies give their employees the ability to warn the public about possible risks without fear of retaliation. They argue that AI technologies can bring significant benefits, but also carry serious risks, such as increased inequality, manipulation and disinformation, and loss of control over autonomous systems.
Justification of the Requirements
Signatories note that companies often hide information about the capabilities and risks of their systems, with minimal obligations to provide data to authorities and no accountability to civil society. In the absence of effective government oversight, employees are among the few people who can hold companies accountable to society.
Basic principles
The letter outlines four key principles that the authors believe AI companies should adhere to:
1. Avoid agreements that limit the right to criticize: Companies should not use non-disclosure agreements (NDAs) to limit the ability of employees to criticize AI technologies.
2. Facilitate an anonymous process: Companies should maintain anonymous channels for employees who wish to report identified risks.
3. Culture of Reasonable Criticism: It is important to maintain a culture in which criticism is accepted as part of the process of improving technology.
4. Avoid retaliating against employees: Companies should not retaliate against employees for disclosing information about AI risks.
The open letter was signed by such well-known figures as former OpenAI employees Jacob Hilton, William Saunders, Carroll Wainwright, as well as Google DeepMind specialists Ramana Kumar and Neil Nanda. These experts expressed deep concern about the current situation in the AI industry.
The situation at OpenAI
It was previously reported that OpenAI forced employees to choose between keeping company shares and the opportunity to publicly criticize the startup. OpenAI CEO Sam Altman confirmed the existence of such a clause, but said the company has never taken away shares and has no plans to do so unless a person signs an NDA or refuses not to criticize the startup.
Experts' message to the AI industry highlights the importance of transparency and openness regarding the risks of AI technologies. The letter calls on companies to create an environment where employees can freely voice their concerns, allowing society to better understand and control the development of AI.
