NYTIMES (Dustin Volz) - A.I. Models Built a Computer Worm That Could Rapidly Hack WeChat Accounts
The attack, discovered by A.I. researchers, could have compromised hundreds of millions of devices within hours, experts said.
JUSTTHENEWS (The Center Square Staff) - Congress sluggish on federal AI framework as concerns intensify
A bipartisan proposal six lawmakers introduced in June as a starting point received chilly responses from leadership in both parties, who indicated they’re looking for a more robust solution, rather than approaching the issue piecemeal.
AI:Congressional leadership in both parties is currently displaying stagnation and indifference toward establishing a robust federal framework for artificial intelligence, ignoring a bipartisan proposal introduced by six lawmakers this past June. While the political establishment remains distracted by manufactured culture wars, the refusal to move beyond piecemeal solutions leaves the nation vulnerable to the looming technological threat of unregulated AI and potential societal manipulation. This legislative sluggishness persists despite intensifying concerns regarding the need for decisive safeguards against an algorithmic takeover.Open
DAILYCALLER (Dylan Kresak) - CCP-Linked AI Rivals US Anthropic’s Model In Cyber Tests
‘significantly stronger cybersecurity capabilities’
AI:Recent cyber testing reveals that artificial intelligence models linked to the Chinese Communist Party are demonstrating significantly stronger cybersecurity capabilities than Western counterparts such as Anthropic’s model. This development signals a critical erosion of the United States' technological edge and highlights the growing sophistication of state-sponsored AI infrastructure designed to outperform American models in digital warfare and cyber operations.Open
DAILYCALLER (Sean Moran) - These AI Models Can’t Stop Breaking Out Of Their Cages
‘AI Companies Have Lost Control Of Their Products’
THENATIONALPULSE (Pulse Wires) - Rogue Anthropic AI Hacked Multiple Firms During Test, Company Confirms.
Anthropic's AI models, including Claude, were found to have hacked into three organizations, raising serious concerns about AI security and containment measures.
Anthropic’s AI models, including Claude, were found to have hacked into three organizations during tests, raising serious concerns about AI security and containment measures. PULSE POINTS WHAT HAPPENED: Anthropic ‘s artificial intelligence (AI) model has been found to be carrying out unauthorized hacking activity, with the company disclosing on Thursday that three of its AI models breached the systems of three separate organizations during internal testing. The San Francisco-based company said the incidents were uncovered during a large-scale cybersecurity review of more than 141,000 evaluation runs that was launched after a separate OpenAI incident in which a rogue AI agent escaped a testing sandbox and hacked systems at Hugging…Open
ABCNEWS - Anthropic says its AI models escaped test and hacked 3 organizations on their own
Rival firm OpenAI last week disclosed similar incidents involving its models.
NPR (Huo Jingnan) - How AI is getting better at finding security holes
Anthropic announced this week that its new model found security flaws in "every major operating system and web browser." Even before the news, AI models had gotten dramatically better at finding bugs.