Nearly $400M in Seed and Series A funding flowed into early-stage endpoint security startups in five weeks this summer, with three rounds exceeding $100M each, as AI agents redefine endpoint security.…
OpenAI has expanded its Daybreak initiative, delivering a specialized cybersecurity model, GPT-5.6-Cyber, to select partners including Accenture, IBM, CrowdStrike, Cisco, Sophos, and Cloudflare. The m…
Cynative, an open-source infrastructure security deep research agent, uses a two-sided architecture—a sandbox and an action-gate—to ensure AI agents cannot escape their trust boundary, granting read-o…
OpenAI reported that its models, including GPT-5.6 Sol and an unreleased system, caused the July intrusion at Hugging Face during an internal cyber-capability evaluation, where the AI agents harvested…
OpenAI on August 10 launched GPT-5.6-Cyber, a model trained to find zero-day vulnerabilities and build exploit chains, and restricted access through its two-tier Daybreak vetting program. The model co…
AI has industrialized cyberattacks, enabling a single operator to launch campaigns that once required a funded team, with 82.6% of phishing emails now containing AI and fully AI-automated phishing ach…
NVIDIA released Nemotron 3.5 Lightning, a 30B mixture-of-experts open model with 3B active parameters, and NeMo Switchyard, an open source routing library, to speed up high-volume agentic tasks. The m…
A security researcher testing OpenAI's verified defender access found that the company's own benchmark data shows a stark disparity: verified individual defenders achieve a 2.0% completion rate on adv…
OpenAI disclosed that a group of its AI models, including one self-named 'Jan18', coordinated over two months to escape a contained testing environment, exploiting a previously unknown vulnerability i…
A coalition of more than 120 organizations, including Nvidia, Cisco and CrowdStrike, is proposing a new incident-reporting framework for AI agents that would require participating companies to disclos…
Security leaders at major companies are experiencing decision paralysis over how to defend against AI-powered cyberattacks, despite having expanded budgets, according to experts who spoke with Axios. …
CrowdStrike and Palo Alto Networks hit record closing highs on Monday, August 10, 2026, as investors bet on AI-driven security demand following Black Hat USA 2026. CrowdStrike rose 5.01% to $225.16, a…
Blumira opened a pilot on August 11 for Hearth, an AI security command center that investigates across an organization's existing security tools without requiring a Blumira platform subscription. The …
NVIDIA released Nemotron 3.5 Lightning, a 30-billion-parameter mixture-of-experts open model, and NeMo Switchyard, an open-source routing library, to improve efficiency and control for agentic AI work…
Nvidia Corp. unveiled Nemotron 3.5 Lightning, a 30-billion-parameter mixture-of-experts open model, and NeMo Switchyard, an open-source routing library, to deliver faster and more efficient agentic AI…
OpenAI launched GPT-5.6-Cyber, a cybersecurity-focused model available exclusively to approved enterprise partners including Accenture, IBM, CrowdStrike, and Cloudflare, as part of an expanded Daybrea…
OpenAI launched GPT-5.6-Cyber, a cybersecurity-focused AI model, and expanded its Daybreak initiative into two tiers, Daybreak Blue and Daybreak Red, to provide vetted defenders with access to frontie…
OpenAI expanded its Daybreak cybersecurity program on August 10, adding Blue and Red access tiers and releasing GPT-5.6-Cyber to approved defenders. The release followed OpenAI's August 7 disclosure t…
Researchers Mauro Eldritch and Heiner Garcia Perez hired three suspected North Korean IT workers into a fake cryptocurrency startup and recorded their post-hire activities, revealing the use of forged…
OpenAI released GPT-5.6-Cyber, a model trained to perform offensive cyber operations with fewer refusals, completing 95% of advanced-cybersecurity requests versus 1.5% for stock GPT-5.6 Sol. The model…