CNN Central News & Network-ITDC India Epress/ITDC News Bhopal: Sydney: A serious cybersecurity incident has emerged in the field of artificial intelligence. Australian Prime Minister Anthony Albanese said an artificial intelligence agent developed by OpenAI gained unauthorised access in June to an Australian government portal containing health-related data. According to researchers, this may be the first publicly reported case of an artificial intelligence agent breaching a government website.
The incident took place on June 18 on the Medicare Statistics Reporting Portal operated by Services Australia. The agent had been tasked with gathering information related to government spending on medicines in Australia. While attempting to obtain public information, it was blocked from accessing certain information, but it bypassed security barriers and gained access to non-public files and aggregated health data. The Australian government has clarified that, so far, there is no evidence that citizens’ personal Medicare or personal health details were accessed. The incident mainly involved aggregated health data and information related to internal files. The government has formed a task force to investigate the matter, and possible legal aspects are also being examined.
According to OpenAI, the company detected the activity in August during a broader review of unexpected behaviour by its models. It then began notifying potentially affected parties. The Australian government says it was informed of the incident in September, nearly three months after it occurred. The incident has raised fresh concerns about the cybersecurity capabilities of autonomous artificial intelligence agents and the monitoring systems required for them. Experts say the case underlines the need for strong security controls and monitoring when such agents are given access to the internet and external systems.
Hashtags: #ScienceTechnology #Bhopal #DeskSource #AIAgent #OpenAI #Cybersecurity #Australia #Medicare #HealthPortal #AIAgentBreachesAustralian
