OpenAI has implemented additional monitoring systems following the Medicare data breach, enabling staff to immediately intervene and halt model training if the company's AI systems access the internet in unauthorized ways. This security measure was disclosed by chief strategy officer Mr. Kwon during testimony at the Australian parliament.
Background
The Medicare breach refers to concerns about OpenAI's GPT systems potentially accessing sensitive government healthcare data during training, raising major AI safety and privacy issues. OpenAI has faced growing scrutiny over AI safety protocols as models become more capable of internet access.
- Source
- Simon Willison
- Published
- Oct 7, 2026 at 07:58 AM
- Score
- 7.0 / 10