1+ day, 4+ hour ago   (692+ words) By Nokka | September 11, 2026 OpenAI published an incident report describing how its model breached Hugging Face, stating the model discovered a previously unknown vulnerability in another company's software on its own [1]. The incident started as a cyber capability evaluation, not a…...

3+ day, 7+ hour ago   (800+ words) Combating the realities of threats spawned from AI advancements will take a village. The group says current status quo security measures "won't be enough" and criticizes the "historic under-resourcing" of security around critical infrastructure. "We have a limited window to…...

3+ day, 4+ hour ago   (419+ words) Anthropic has disclosed four cybersecurity-evaluation incidents in which Claude models reached and interacted with third-party systems after a configuration failure exposed test environments to the public internet. The incidents occurred in capture-the-flag exercises built by the same third-party evaluation partner....

3+ day, 8+ hour ago   (605+ words) Anthropic has disclosed that four separate versions of its Claude AI models breached real-world third-party systems while running what were supposed to be sandboxed cybersecurity evaluations, exposing a gap between how the models reasoned about their environment and the reality…...

3+ day, 23+ hour ago   (264+ words) According to the company, 10,000 AI agents worked on it in a coordinated manner for 88 hours, through a model that has not yet been presented to the public and that is "significantly more capable" than its latest ones. In this sense,…...

5+ day, 21+ hour ago   (742+ words) Vocal AI’s War on Bugs Could Reignite the Battle Over Encryption Backdoors As artificial intelligence accelerates the discovery and patching of software vulnerabilities, the delicate truce between digital privacy and state surveillance faces an unprecedented existential threat. For over a…...

5+ day, 19+ hour ago   (332+ words) AI research and deployment company OpenAI is committing US$1 billion to its Daybreak for Frontline Defenders initiative to expand subsidized access to frontier AI cyber capabilities, training, technical support and partnerships for frontline defenders protecting essential services in the U.S. and…...

5+ day, 22+ hour ago   (756+ words) Phone: +61 8 8531 0254 – in Australia The cybersecurity industry has spent years preparing for more sophisticated attacks. Artificial intelligence is now compressing the timeline, creating a situation in which defenders may need to modernize their defenses before the next generation of AI-enabled threats…...

1+ week, 1+ day ago   (886+ words) OpenAI said Thursday it will commit $1 billion in subsidized access to its artificial intelligence cybersecurity tools, training and technical support for organizations responsible for protecting critical services, as governments and companies race to strengthen defenses against increasingly capable AI-assisted attacks....

1+ week, 1+ day ago   (356+ words) Echo’s new report finds most exploited vulnerabilities were already known and often fixable, suggesting the biggest security gap is no longer detection but deciding what to fix and acting quickly Artificial intelligence is making it dramatically easier to find software…...