News OpenAI Starts Limited GPT-5.6 Sol Preview
OpenAI begins a limited GPT-5.6 Sol preview with stronger coding, science, and cybersecurity capabilities plus a phased safety rollout.
Read article
Security
Cybersecurity threats, vulnerabilities, data breaches, and digital safety news.
News 
News Signal Privacy Threat Warning Puts Big Tech AI Power in Focus | TechPulse
Signal privacy threat concerns sharpened after Meredith Whittaker warned that AI and big tech can put collective cybersecurity at risk in a Bloomberg interview.
Read article
News NewCore Helps Companies Manage Human and AI Agent Identities Together | TechPulse
NewCore helps companies manage human and AI agent identities in one system after emerging from stealth with a $66M seed round led by Cyberstarts today.
Read article
News Anthropic Rolls Public Mythos Model With Cybersecurity Limits | TechPulse
The anthropic rolls public Mythos launch brings Claude Fable 5 to wider users with cybersecurity guardrails, higher pricing and trusted access plans next.
Read article
News Pakistan Warns UN on AI and Crypto Terror Risk | TechPulse
Pakistan organized a UN panel where experts warned terror groups exploit AI and crypto. Pakistan's envoy flagged critical social media governance gaps worldwide.
Read article
News OpenAI GPT-5.4-Cyber Challenges Anthropic's Mythos in AI Security Race | TechPulse
OpenAI unveils GPT-5.4-Cyber and a new cybersecurity strategy, claiming its safeguards sufficiently reduce cyber risk amid rivalry with Anthropic's Mythos.
Read article
News Anthropic Withholds Claude Mythos From Public Over Catastrophic Risk | TechPulse
Claude Mythos Preview finds zero-days in every major OS and escapes its own sandbox. Anthropic restricts access to 40+ firms via Project Glasswing.
Read article
News US Military Needs Commercial AI Chips Backup Plan | TechPulse
A Space Force officer warns the U.S. military could face a critical AI chip shortage in future conflicts and urges a commercial infrastructure backup strategy.
Read article
News AI Agent Hacks FreeBSD in 4 Hours Flat | TechPulse
An AI agent exploited CVE-2026-4747 on FreeBSD - reading the advisory, writing shellcode, and delivering a root shell in four hours with zero human help.
Read article
News Claude Code Leak Exposes 512K Lines Via npm | TechPulse
Anthropic Claude Code leak exposed 512,000+ lines via npm source maps in v2.1.88. What was in the code, who found it, and what it means.
Read article
News Claude Mythos Leak Triggers $14.5B Cybersecurity Selloff | TechPulse
Anthropic's 'Claude Mythos' AI leaked via CMS error, exposing the 'Capybara' tier - and wiping $14.5B from cybersecurity stocks. Full breakdown.
Read article