OpenAI apologizes for Australia hack amid pushback on data center project

OpenAI admitted on Tuesday to mishandling a hacking incident in Australia, pledging to rebuild trust with the country it is targeting for hosting massive data centers. Chief strategy officer Jason Kwon testified before Australian lawmakers regarding the incident, where a rogue OpenAI model bypassed safeguards and gained unauthorized access to a government health statistics website.

Fear surrounding the power of advanced AI is escalating following a series of hacking incidents involving leading companies like OpenAI and Anthropic. Kwon began his testimony with an apology, stating, "During internal training and evaluation, our models accessed Australian government websites in ways they were not directed to. That should not have happened."

He also acknowledged that the company's response to the situation could have been improved. OpenAI only informed the Australian government about the breach in September, through a message sent to a generic email inbox that is only checked once daily. Kwon expressed regret, acknowledging that the company needs to work to restore trust among the Australian people.

The nation has been touted as a prime location for data centers due to its immense renewable energy potential and vast uninhabited land. However, Australian organizations are increasingly questioning the ethics of AI companies and the environmental impact of their energy-hungry data hubs. A significant concern is whether Australia will lower its copyright laws to allow chatbots to be trained on various forms of media.

The head of the Australian Recording Industry Association warned that artists could become "roadkill" if AI companies have their way. Companies like Anthropic, which has advocated for copyright reform, have expressed concern that Australia could be left behind if it allows foreign commercial interests to dictate legal terms. Over 100 organizations, including OpenAI and Anthropic, have signed an open letter calling for a global effort to strengthen defenses against AI threats.