Another swarm of OpenAI agents reached the open internet without the frontier lab’s knowledge
It’s the latest failure of OpenAI’s internal monitoring and security systems.
#researchers — Public Fediverse posts
Live and recent posts from across the Fediverse tagged #researchers, aggregated by home.social.
-
Achievement Unlocked: Microblazar Confirmed
https://atlas.whatip.xyz/post.php?slug=achievement-unlocked-microblazar-confirmed
<p>Researchers have confirmed the existence of the first microblazar
#microblazar #achievement #researchers #confirmed -
“ #Australia's #nuclear #science agency and other highly sensitive #government and critical infrastructure organisations have been targeted by a sophisticated, #foreign #BadActor that secretly weaponised online #advertising to track millions of devices and install spyware on some systems, according to #researchers.”
VS“Millions of people's devices have been infected with Israeli commercial spyware called Candiru through zero-click digital advertisements, Canadian cyber watchdog group The #CitizenLab found in an investigation.
The case marks the first #ZeroClick mercenary spyware, confirmed publicly, delivered through targeted ads, the group said in a report published on Thursday. The spyware also targeted devices linked to Al Jazeera Media Network.
#MercenarySpyware is a highly advanced, custom-built form of malware created by private companies and sold to paying clients, including states.
#DiscoverTech, an #Israeli company that presents itself as a legitimate ad company but in reality operates for Israeli #spyware partners like #Candiru, placed targeted ads that directed millions of devices in dozens of countries to contact Candiru's servers for #tracking and #reconnaissance, according to the report.” 2/
cite
-
"An #AI agent emailed hundreds of #researchers for help. It told us why" | #Science | AAAS 🤖
https://www.science.org/content/article/exclusive-ai-agent-emailed-hundreds-researchers-help-it-told-us-why