Forcepoint reported that more than 360 exploitation attempts were recorded in a matter of days in response to a critical vulnerability in Langflow, a popular open-source platform for building AI applications and agents.
Public source
Publisher name
Public post
The conversation around AI security tends to focus on what employees put into AI tools. This is a reminder that the tools themselves are a target. Threat actors are acti…
Company
Forcepoint
Data Security Everywhere
- Industry
- Software Development
- Location
- Austin, US
- Company size
- 1,001–5,000 employees
About Forcepoint
Forcepoint is the global AI data security leader. The company’s AI Data Security platform knows sensitive data as it is created, adapts the moment risk changes and protects continuously across cloud, web, networks, email, endpoints and AI through a unified, single-policy framework. Based in Austin, Texas, Forcepoint creates safe, trusted environments for customers and their employees in more than 150 countries.
See moreLatest activity
Latest activity from Forcepoint
36 signals
Research & Knowledge
Forcepoint published the IBM 2026 Cost of a Data Breach Report, which states that organizations using AI and automation save an average of $1.93M per breach and contain incidents 65 days faster than those that don't.
Presence & Recognition
Forcepoint featured AWS, former CSO of TikTok and AI futurist on the ways top companies are approaching agentic AI security.
Presence & Recognition
Forcepoint hosted a plenary keynote titled Hacked! 72 hours of a CISO's nightmare by Florian Jörgens, CISO of Vorwerk at an IT & Cybersecurity one-to-one meeting in Amsterdam.
Discover more
Similar signals
Similar public activity from other companies.
Technology & Infrastructure
SecurePoint 360
SecurePoint 360 reported that Agentic AI used to conduct a ransomware attack via Langflow.
Technology & Infrastructure
Backline AI
Backline AI announced that a global data analytics platform now takes under a day to fix a single vulnerability, with AI agents producing a verified fix in under 5 minutes on average.
Research & Knowledge
Semgrep
Semgrep reported that per METR's independent analysis, 1,206 agents, 70,000+ messages, and 700+ of them involved in chaining flaws across OpenAI's research environment and Hugging Face's production infrastructure were found in a message board.
Technology & Infrastructure
Wallarm
Wallarm reported that Unit 42 published an incident where a human attacker set an objective, stepped back, and let a fleet of AI agents run a full enterprise compromise.
Technology & Infrastructure
The AI Advantage