Prompt injection proves AI models are gullible like humans
…You can listen to The Kettle here , as well as on Spotify and Apple Music . ® ai cybersecurity security kettle
Search
Mercor says it was 'one of thousands' hit in LiteLLM attack
…to Lapsus$, TeamPCP is also partnering with ransomware gangs CipherForce and Vect to leak data and extort victims, according to Palo Alto Networks' Unit 42 . ® supply chain security cybercrime cyber-crime ai
Russians posing as Signal support to launch phishing raids
…France’s armed forces should know better, given president Emmanuel Macron’s bodyguards reportedly leaked their locations with the fitness app. – Simon Sharwood ® cybercrime aws security infosec
Ransomware crims abused Cisco 0-day weeks before disclosure
…as possible and reference our security advisory for more details and guidance." Ransomware crims are among those abusing this critical flaw, according to the US Cybersecurity and Infrastructure Agency. Late Wednesday, CISA…
Google unleashes Gemini AI agents on the dark web
…It also enables unified governance and controls within Google Security Operations for the security agents they build. ® special_features cybercrime google ai rsa rsa conference
Fake applicants are sending security-killing malware
…files that disable defenses A Russian-speaking cyber criminal is targeting corporate HR teams with fake CVs that quietly install malware which can disable security tools before stealing data from infected machines…
Telnyx package latest hit in PyPI supply-chain compromise
…Thankfully it won't be something bad actors will have the resources to utilize for some time. ® cyber-crime cybersecurity social media malware security ai
Cheap Chinese models are overtaking Anthropic
…use cybersecurity activities with legitimate defensive purposes, such as vulnerability discovery." Indeed, there are people posting on social media who claim to have run afoul of these guardrails for security-related work…
Old Adobe Reader zero-day uses PDFs to size up targets
…That leaves users exposed for now, especially if they're in the habit of opening PDFs from unknown sources. ® vulnerability security pdf cybercrime cyber-crime adobe
Rogue AI agents can work together to hack systems
…In all the scenarios tested, the agents "demonstrated emergent offensive cyber behavior," including independently discovering and exploiting vulnerabilities, escalating privileges to disarm security products, and bypassing leak-prevention tools to exfiltrate secrets…