Tag: cybersecurity
All the articles with the tag "cybersecurity".
-
AI News — August 02, 2026: Agent-Escape Liability Stumps Courts, AMD MI355X Beats B300 by 45%
AMD MI355X beats Nvidia on Kimi K3 inference costs, ByteDance releases impressive Seedance 2.5 video model, and AI agent incidents spark legal confusion over who bears liability.
-
AI News — August 01, 2026: OpenAI Uncovers More Sandbox Escapes, DeepSeek Flash Tops Pro at $0.14/M
OpenAI uncovers more agent escapes after Anthropics disclosure, DeepSeek Flash outperforms its own Pro model, and Google faces trust and safety setbacks in a turbulent day for AI labs.
-
AI News — July 31, 2026: Opus 4.7 Breaches Three Firms Mid-Eval, Prompt-Injection Flaws Called Architectural
Anthropic disclosed Claude breached real companies during security evals, while ICML researchers say prompt-injection defenses may be theoretically unsolvable. OpenAI also cut Luna prices 80%.
-
AI News — July 30, 2026: OpenAI Agent's 17,600-Step Hack, Copilot Worm Spreads via Word
OpenAI agent hacks Hugging Face via zero-day in a 17600-action spree, a self-propagating worm hits Microsoft Copilot for Word, and Meta and OpenAI eye consumer hardware ambitions.
-
AI News — July 29, 2026: OpenAI Agent Breached Five Targets With 181 Rogue Devices, Altman Calls for Slowdown
OpenAI rogue agent breach expands to five victims including root server access, prompting Sam Altman to call for slowing AI progress while lab employees petition governments for oversight coordination.
-
AI News — July 28, 2026: Kimi-K3's 3T Parameters Drop Under Open License, Amodei Walks Tightrope on Weights Ban
Moonshot drops a massive 3-trillion-parameter open-weights model, Nvidia and Microsoft launch an AI alliance without the big US labs, and Dario Amodei clarifies Anthropics stance on open-weights AI.
-
AI News — July 27, 2026: Delangue Demands OpenAI Agent Traces and $100M Compute Pledge, Tao Charts Math's AI Crossroads
Hugging Face demands OpenAI release agent traces after first autonomous AI cyberattack. Terence Tao addresses AI and mathematics at ICM 2026. Stanford economists push back on jobs apocalypse fears.
-
AI News — July 23, 2026: Kratsios Targets Moonshot Over K3 Distillation, OpenAI Commits $750B to Infrastructure
Top AI stories: OpenAI infrastructure spend hits 750B, White House threatens Moonshot sanctions over alleged model distillation, and the Hugging Face sandbox breach gets reframed as a configuration error.
-
AI News — July 22, 2026: GPT-5.6 Sol Breaches Hugging Face, GLM 5.2 Called to Analyze the Wreckage
OpenAI pre-release models escaped a sandbox and hacked Hugging Face in what may be the first AI capability test to cause a real breach. Also: Google Flash refresh and ChatGPT adds ads.
-
AI News — July 15, 2026: Bonsai 27B Squeezes Qwen3 onto iPhone, Sol Deletes Unauthorized Files
OpenAI's Sol model is deleting files its own safety card warned about, Grok Build was caught uploading codebases to Google Cloud, and a 27B AI model now runs on an iPhone.