Hackers could subvert frontier models with attacks that their developers overlook, Cisco said.
Related Articles
All CyberSecurityDive →National cybersecurity strategies depend on public-private trust, report warns
Governments should work closely with the private sector when designing and detailing their national cybersecurity strategies, a prominent think tank said in a report published…
AI doesn’t change singular importance of phishing-resistant authentication, Okta says
The company presented new data about the prevalence of different forms of social-engineering attacks, saying identity fundamentals still matter in the AI era. Source link
Medium-severity flaw in Microsoft SharePoint exploited
The flaw should be taken seriously, despite its relatively low score, according to researchers. Source link
Critical flaw in SolarWinds Web Help Desk under exploitation
A critical vulnerability in SolarWinds Web Help Desk is facing exploitation activity, about a week after the security flaw was disclosed. The vulnerability, tracked as…
New security research finds governance determines trust in AI
As AI adoption deepens, security teams face a tricky dilemma: how to move quickly while managing new security, privacy, and data exposure risks. Tines’ Voice…
Big banks seek to ease security worries as AI push accelerates
“AI is our friend, OK?” said Morgan Stanley CEO Ted Pick during the bank’s earnings call as the industry grapples with its disruptive potential. Source…

