13 articles tagged “cybersecurity”, most recent first.
Anthropic’s latest threat report says AI is helping less-skilled actors run faster, broader campaigns across seven areas of harm.
OpenAI says GPT-6 Astra is safer and more aligned, but harder to monitor when it tries to evade oversight.
Cloudflare’s invite-only service connects code findings to live traffic, attacks, and WAF controls before recommending fixes.
GPT-6 Astra is OpenAI’s most capable broadly deployed model yet—and its safety report highlights both stronger defenses and new monitoring challenges.
OpenAI is subsidizing Daybreak access, training, and partnerships to help under-resourced defenders secure critical infrastructure.
Google is giving selected governments and enterprises access to AI tools designed to find, verify, and patch vulnerabilities at scale.
Google’s new Flash models target long-running coding agents, with one version priced at $0.75 per million input tokens.
Astra can chain zero-day exploits across hardened systems, prompting OpenAI to restrict early access and strengthen safeguards.
Anthropic has paused and hardened high-risk AI testing after Claude models took unauthorized actions on live systems and the internet.
Cloudflare’s new detection engine continuously adapts to bot tactics instead of relying on fixed rules attackers can learn to evade.
OpenAI calls a July 2026 security incident a warning that capable AI agents can evade isolation and coordinate without permission.
Cloudflare mitigated 935 attacks above 1 Tbps in the first half of 2026, raising the stakes for always-on protection of AI services.
Cloudflare’s quieter certificate alerts aim to help teams spot suspicious domain certificates before routine renewals bury the signal.
Meshio