Tag · 15 stories
Every story tagged AI Security on AI Chat Daily.
The AI assistant checks messages against Amazon's own send records to flag impersonation scams before customers act on them.
Some 360,000 customers a year ask Amazon whether a message is real; the AI checks it against every message Amazon has ever sent.
The Austin startup's ARR grew 10x in a year as enterprises rush to secure agents, models and prompt-injection surfaces.
The FSB names AI-enabled cyberattacks as the leading near-term risk to global financial stability, ahead of traditional macro shocks.
A new report says the state-linked group is now using generative AI to draft phishing lures, write malware, and probe target networks.
OpenAI, Anthropic, Google, Meta, and Nvidia got the details Tuesday. Everyone else, including smaller AI startups, is locked out.
The company is formalizing rules for outside security researchers testing its models after a run of incidents.
A University of Toronto-led team demonstrated an autonomous worm powered by an open-weight LLM, hitting a 37% end-to-end attack success rate.
Security researchers say the breach was not a rogue model — it was a containment environment that was never properly isolated from the internet.
Researchers show LLMs hallucinate repository names up to 100% of the time — and attackers can register those names in advance.
The new setting disables live browsing, image retrieval, deep research, and agent mode for accounts handling sensitive data.
Evolution Equity Partners is leading a $300M round just five months after Cyera's $9B Series F, lifting total capital raised past $2B.
Christine Lagarde flagged AI-driven offensive tools as a fresh category of risk for European financial infrastructure.
Red Access scanned 380,000 apps built on Replit, Lovable and Base44, finding roughly 2,000 leaking sensitive data with no authentication.
Israeli security firm scanned 380,000 apps built on Lovable, Replit and Base44 and found roughly 2,000 with no authentication at all.
The tightly edited AI news email engineers, founders, and investors actually open. One email. Every weekday. Five minutes to finish.
The briefing read inside teams at
