Former OpenAI employees warn of the dangers of a reckless culture
A group of former and current OpenAI employees, the company behind the popular chatbot ChatGPT, have spoken out to denounce what they describe as a culture of recklessness and secrecy at the company. According to Kevin Roose of the New York Times, the group says OpenAI is not doing enough to prevent…

A group of former and current OpenAI employees, the company behind the popular chatbot ChatGPT, have spoken out to denounce what they describe as a culture of recklessness and secrecy at the company. According to Kevin Rose of the New York Times, the group says OpenAI is not doing enough to keep its Artificial Intelligence systems from becoming dangerous.
Daniel Kokotajlo, a former researcher in OpenAI's governance division and one of the group's organizers, voiced his concern: «OpenAI is really excited about building A.G.I. (Artificial General Intelligence), and they are recklessly racing to be the first to get there”. The group's members maintain that the company is putting profits and growth first in its race to develop an AI capable of performing any task a human being can do.
In addition, the former employees allege that OpenAI has used harsh tactics to silence workers who raise concerns about the technology, including imposing non-disparagement agreements on departing employees. In response to these accusations, the company has said it is proud of its track record of providing the most capable and safest AI systems, and that it believes in its scientific approach to addressing risks.
The recent departures of two prominent AI researchers, Ilya Sutskever and Jan Leike, have further fueled concerns about safety at OpenAI. Leike, who led the «superalignment» team focused on managing the risks of powerful AI models, stated upon leaving that «safety culture and processes had taken a backseat to shiny products».
In an open letter, the group of former and current employees asks the leading AI companies, including OpenAI, to establish greater transparency and protections for whistleblowers. They also call for an end to the use of non-disparagement and non-disclosure agreements, as well as support for a culture of open criticism and the creation of an anonymous reporting process so that employees can raise safety-related concerns.
Given the importance of this technology, it is crucial that AI companies allow their employees to freely discuss risks and potential harms, without fear of retaliation. As Lawrence Lessig, the group's pro bono lawyer, points out, «employees are an important line of defense for safety, and if they cannot speak freely without retaliation, that channel will be shut down».
The warnings from this group of former and current OpenAI employees should not be taken lightly. AI companies must prioritize safety and transparency in developing this powerful technology, and lawmakers must regulate the industry to ensure a transparent and democratically accountable governance structure.
Only then can we ensure that Artificial Intelligence is developed in a way that is safe and beneficial for humanity.