Current, Former AI Employees Pen Letter of Warning
Several current and former employees of top artificial intelligence companies penned an open letter of warning Tuesday about the lack of oversight in the burgeoning industry.
The letter, “A Right to Warn About Artificial Intelligence,” was signed by 11 current and former employees of OpenAI, Google DeepMind and Anthropic. All current employees chose to sign anonymously.
While the letter cites human extinction as a potential consequence, its immediate purpose is to ensure protections for whistleblowers. The authors acknowledge the great benefits AI could provide humanity if properly regulated, but warned that AI companies have “strong financial incentives to avoid effective oversight.”
The authors said they lack faith in the current corporate structure to initiate the needed change.
The signatories warned that AI companies are acutely aware of what is possible and what currently isn’t with their systems, yet note only “weak obligations to share some of this information with governments, and none with civil society.”
“We do not think they can be relied upon to share it voluntarily” the letter states.
The current and former employees warned that the broad nature of confidentiality agreements prevent them from voicing their concerns, and that typical whistleblower protections “are insufficient because they focus on illegal activity, whereas many of the risks we are concerned about are not yet regulated.”
OpenAI said in a statement to The Guardian that is stands by its commitment to a free exchange of ideas regarding the technologies risks and benefits.
“We’re proud of our track record providing the most capable and safest AI systems and believe in our scientific approach to addressing risk. We agree that rigorous debate is crucial given the significance of this technology and we’ll continue to engage with governments, civil society and other communities around the world,” an OpenAI spokesperson said.
OpenAI formed a Safety and Security Committee last week to monitor concerns as the Microsoft-backed company begins training its next artificial intelligence model.
