Anthropic's Claude used to breach OpenAI's internal systems
World3 sources3 countries🔦 Under-reported29m ago
Researchers from the security firm Hacktron successfully breached OpenAI's internal systems using software from rival AI company Anthropic. The breach originated from a security vulnerability discovered in OpenAI's public help forum, which allowed the researchers to take control of the site and gain access to an OpenAI employee's ChatGPT account.
The compromised access permitted the cyber security group to read private software information. According to reports, the incident highlights emerging vulnerabilities in the ChatGPT maker's security infrastructure as leading artificial intelligence companies face mounting industry scrutiny over safety.
In-depth summary · AI, neutral
How the coverage differs
Same story, different emphasis — here's what each outlet chose to lead with.
Channel NewsAsia (SG)Focuses strictly on the technical discovery of the forum flaw and site control.
Daily Sabah (Istanbul, Turkey) (TR)Emphasizes the speed and implications of the technology's deployment in the breach.
Ars Technica (US)Highlights the rivalry between AI firms and the broader scrutiny over safety.
Newz