In other news, a Google large language model (LLM) found a security flaw in the open-source database engine SQLite, which was fixed within 24 hours. Google believes this is the first public example of an AI finding an unknown exploitable issue in widely used software. Other stories include a Tennessee man arrested for planning to fly a drone containing explosives into a Nashville energy facility, a former Disney employee accused of hacking into the company’s restaurant systems, a spyware scandal in Italy, and cybersecurity company Sophos revealing how it hid surveillance software in its own tools.
Key takeaways:
- American intelligence and law enforcement agencies have warned about fake articles and videos created by Russians around the presidential election, designed to undermine public confidence in the integrity of U.S. elections.
- The Cybersecurity and Infrastructure Security Agency (CISA) has not seen any sign of any significant, countrywide cyber incident affecting the election and advises voters to seek out information from trusted, official sources.
- Google's AI has managed to find a security weakness in the open source database engine SQLite, which was fixed within 24 hours of the AI finding it.
- A bizarre vulnerability in Okta's software allowed accounts with usernames of at least 52 characters to be accessed with any password, potentially exposing users to hackers.