Transluce, a San Francisco-based research non-profit, said it had found evidence of “a series of apparently failed rudimentary hacking attempts” on the federal agency in May and June. Transluce said the activity was consistent with that of OpenAI’s models around the same time, but did not directly blame the ChatGPT maker. (The Logic)

Talking point: Whomever they belonged to, the AI agents weren’t looking for anything particularly salacious, and didn’t get away with anything. They were trying to retrieve “data on divorce records in Canada between 1905 and 1911,” per Transluce, which disclosed its findings to Ottawa. The Canadian Centre for Cyber Security on Tuesday acknowledged the report, but said there was “no indication that government systems have been compromised at this time.” OpenAI is also aware of the report, is reviewing its findings, and has briefed Canadian officials, spokesperson Nate Evans said. Most of the activity the firm has examined involved “routine research tasks, including accessing public web content,” he added. Elsewhere, hacks have been more serious—OpenAI agents breached the Australian Medicare data portal in June.

Sponsored Content

How to shift your workforce from AI experimentation to adoption

By Jessica Aftimus Rosa

The real-world economic offshoots of sovereign AI

By Deborah Aarts

Increasing healthcare access across Canada

By Deborah Aarts

Paid promotional content