Skip to content
NEAR AI Breach Exposes Data of 100 Users via SGLang Vulnerability

NEAR AI Breach Exposes Data of 100 Users via SGLang Vulnerability

First seen 10 Nov 2026, 20:21 UTC • •

Article Content

Browse articles
ThreatCluster AI
ThreatCluster •November 11, 2026 at 19:46 UTC
  • •An attacker exploited a remote code execution vulnerability in SGLang, affecting NEAR AI's infrastructure.
  • •Data from approximately 100 users and 18 AI agents was exposed, including email addresses and chat histories.
  • •NEAR AI has taken steps to mitigate the breach, including revoking access and patching the vulnerability.

NEAR AI reported a security breach on November 9, 2026, where an attacker exploited a remote code execution vulnerability in SGLang, an open-source inference engine. This breach allowed access to the names and email addresses of approximately 100 users, along with some account details and data from 18 AI agents, including chat histories and logs. The attacker gained access to an internal system linked to a deprecated feature, Private Chat. NEAR AI confirmed that no stored Private Chat conversations were accessed, but there is a residual risk regarding live requests processed on November 8. The company has since revoked all active sessions, replaced credentials, and patched the vulnerability. NEAR AI has also reported the flaw to the SGLang developers. The incident highlights the risks associated with vulnerabilities in software running in trusted execution environments.

Start a free Starter trial for enhanced analysis

Ask AI about this cluster

Timeline

2026-11-08
Potential access to confidential requests
The attacker may have observed live requests being processed on the compromised server, including those in confidential mode.
Cryptoticker
2026-11-09
Breach disclosed
NEAR AI published a report detailing the breach and the data exposed, confirming the exploitation of the SGLang vulnerability.
Tech.Yahoo
2026-11-10
Vulnerability patched
NEAR AI patched the SGLang vulnerability across all inference servers and restored access to NEAR AI Cloud.
Near.Ai

More articles in this cluster (6)

Following this threat?

Track NEAR AI in your own feed — you're alerted when they show up in new reporting, leak sites or exploitation.

Free account · no card needed

Common questions

What data was compromised?
The breach exposed names, email addresses, subscription plans, account balances, and data from 18 AI agents.
What actions has NEAR AI taken?
NEAR AI revoked all active sessions, replaced credentials, patched the vulnerability, and notified affected users.
Is there a risk of further exploitation?
While the vulnerability has been patched, there is a residual risk regarding live requests processed during the breach.