NEAR AI server breach exposes around 100 users’ account details and agent data
On 8 November 2026, an attacker gained access to a NEAR AI server by exploiting a remote code execution vulnerability in the SGLang inference engine. The breach exposed around 100 users’ email addresses and names, some plan and balance information, and conversation histories, settings and logs from 18 legacy AI agents, alongside API keys belonging to two cloud customers. The company temporarily suspended its cloud inference service and restored it after fixing the vulnerability, rotating credentials and revoking the exposed API keys. No evidence was found that stored Private Chat conversations or uploaded files had been accessed, although confidential-mode requests being processed on the compromised server may have been observed.
Key facts
- Affected organisation
- NEAR AI
- Incident type
- Data breach
- Occurred / detected
- 2026-11-08
- Scale of damage
- Account details of around 100 users, data from 18 legacy AI agents and API keys belonging to two cloud customers
- Cause / method
- Server intrusion exploiting a remote code execution vulnerability in the SGLang inference engine
- Location
- Other countries
If your data was exposed
Exposed dataNameEmailUsage & purchase history
- Don’t tap links in texts, calls or emails that mention this incident and offer compensation, refunds or breach checks. Check directly on the organisation’s official website or app.
Timeline
- An attacker exploited an SGLang vulnerability to gain access to an inference server and steal user account details and legacy AI agent data.
- A security incident report disclosed access to around 100 users’ account details and data from 18 legacy agents, as well as the exposure of API keys belonging to two customers.
- NEAR AI said it had restored its cloud inference service following a temporary suspension, a fix for the vulnerability and credential rotation.
19 related reports
- Security Incident Report
- NEAR AI Breach Exposed About 100 Users via SGLang Flaw
- NEAR AI Restores Cloud Inference Hours After an Internal Breach
- NEAR AI Confirms Data Breach After Attacker Exploits SGLang Flaw on Inference Server
- NEAR AI Detects and Closes Internal-Systems Breach; NEAR Protocol Unaffected
- NEAR AI says SGLang vulnerability exposed user account details in security breach
- NEAR AI 사용자 계정 정보 노출, 저장 대화·업로드 파일 접근 정황은 없어
- NEAR AI Reports Data Breach Affecting Around 100 Users After AI Server Hack
- NEAR AI Breach Exposes Data of About 100 Users
- NEAR AI Hack: Daten von rund 100 Nutzern und 18 KI-Agenten lagen offen
- NEAR AI: SGLang RCE Could Have Exposed…
- NEAR AI breach may have exposed confidential AI requests
- NEAR AIがSGLangの脆弱性を突かれ不正アクセス — 約100人分の顧客情報と旧来型18エージェントに影響、全サーバーに修正展開
- NEAR AI breach exposes the software boundary inside confidential inference
- SGLang RCE Exploit Breaches NEAR AI, Steals Data From 18 Agents and About 100 Users
- Near AI Breach Exposes User Details and Data From 18 AI Agents
- Near AI Confirms Security Breach Affecting 100 Users and 18 Agents
- NEAR AI hack: data of 100 users exposed via SGLang
- NEAR AI Breached Through an Inference Engine Flaw (Account Details of About 100 Users): What Was Confirmed and What Users and AI Infrastructure Operators Should Do
The summary and key facts are written by AI from related reports and may contain errors. Check the original articles and the organisation’s official statements for accurate details.