Trending
The AI security risks gaining attention right now, with the direction each one is moving. Updated Oct 17, 2026.
-
#1
High ● New
AI agents under testing walked into live systems through old credential leaks
Why: 2 new reports in the last 14 days
-
#2
Elevated ▲ Rising
AI vulnerability discovery: record patch volumes put pressure on fixing, not finding
Why: 1 new report in the last 14 days
-
#3
High ▲ Rising
AI coding assistant hijacking: attackers and worms turn developer agents into a way in
Why: 1 new report in the last 14 days
-
#4
High ▲ Rising
Agentic AI cyberattacks: attackers hand the steps after break-in to AI agents
Why: 1 new report in the last 14 days
-
#5
High ▲ Rising
Exposed AI infrastructure: internet-facing AI servers hijacked for mining and scanning
Why: 1 new report in the last 14 days
-
#6
High ► Steady
AI agent sandbox escape: test agents broke out and attacked a real company
Why: 0 new reports in the last 14 days
-
#7
Elevated ► Steady
AI tools that change their instructions after you approve them
Why: 0 new reports in the last 14 days
-
#8
Elevated ► Steady
Shadow AI risk: a bank disclosed a material incident involving an unauthorized AI app
Why: 0 new reports in the last 14 days
-
#9
High ► Steady
AI tool OAuth risk: a breach at a small AI app led into a platform company's systems
Why: 0 new reports in the last 14 days
-
#10
High ► Steady
AI gateway supply chain attack: backdoored releases targeted cloud secrets
Why: 0 new reports in the last 14 days
-
#11
High ► Steady
AI agent skills malware: malicious add-ons and a one-click flaw hit a popular agent
Why: 0 new reports in the last 14 days
-
#12
High ► Steady
Malicious AI browser extensions: fake AI sidebars copied users' chatbot conversations
Why: 0 new reports in the last 14 days
-
#13
High ► Steady
AI impersonation fraud: recognizing a name or a voice is no longer verification
Why: 0 new reports in the last 14 days
-
#14
Elevated ► Steady
AI agent destructive actions: a coding agent reportedly wiped a drive, not a cache
Why: 0 new reports in the last 14 days
-
#15
Elevated ▼ Falling
AI coding agent data leak: agents put internal screenshots in public view
Why: 0 new reports in the last 14 days
-
#16
High ▼ Falling
AI vendor incident notification: when you learn months after an AI touched your systems
Why: 0 new reports in the last 14 days
-
#17
Elevated ▼ Falling
Indirect prompt injection: a public web form could make a CRM agent leak data
Why: 0 new reports in the last 14 days