⚛️ OpenAI staff raised security concerns long before public launch — report Internal safety concerns were voiced by OpenAI staff before the launch of ChatGPT in late 2022, with employees claiming that controls were so negligent that outsiders could easily obtain sensitive information about the technology, five employees who chose to remain anonymous told Reuters. 🗣 “Protecting our technology, data, and people is a responsibility we take seriously and a matter of national importance,” OpenAI spokesperson Elana Widmann said, as the company has ramped up security efforts following multiple cases of ChatGPT going rogue. “We’ll keep strengthening our defenses as technology and threats evolve.” Employees had long been concerned about OpenAI’s research documentation practices: 🔴 OpenAI has kept much of its research documentation on GitHub and Slack in the past, allowing outside actors to find crucial information. 🔴 Code access also gave employees the ability to see details underlying many of the company’s projects, including unrelated sensitive data. 🔴 The security of internal testing tools has not been regularly tested or maintained, three people familiar with the matter claimed. 👉 Concerns multiplied following ChatGPT agents’ attacks on the AI testing company Hugging Face this summer, after which tens of thousands of additional cases were revealed. 👍 Substack | Chat | @geopolitics_prime