Malware & Threats

Developers Targeted With Malware Disguised as DeepSeek Package

Python developers looking to integrate DeepSeek into their projects were targeted with malicious packages delivered through PyPI.

DeepSeek malware generation

Threat researchers have come across two malicious Python packages offered as resources for integrating the Chinese AI model DeepSeek into software projects.

The malicious packages, named ‘deepseeek’ and ‘deepseekai’, were uploaded to the Python Package Index (PyPI) package repository by a user named ‘bvk’ on January 29. 

The fake DeepSeek packages were detected in minutes by cybersecurity firm Positive Technologies and PyPI administrators removed them within an hour of their publishing. 

However, they were still downloaded more than 200 times before they were removed, including over 100 times from the United States. 

An analysis showed that the fake DeepSeek packages hid malicious functions designed to collect user and system data, as well as environment variables.

“Environment variables often contain sensitive data required for applications to run, for example, API keys for the S3 storage service, database credentials, and permissions to access other infrastructure resources,” Positive Technologies noted.

Advertisement. Scroll to continue reading.

The malware is designed to send the stolen data to a command and control server through the Pipedream integration platform. 

“It’s worth mentioning that the script was written with the help of an AI assistant, which is indicated by the characteristic comments explaining the lines of code,” the security firm said.

The attack was likely aimed at developers, ML engineers or AI enthusiasts, according to Positive Technologies. 

“Cybercriminals always monitor the current trends and will try to take advantage of them at the right moment. In this case, we analyzed a relatively harmless attack, although due to the hype around DeepSeek, there could be a lot more victims if the malicious package activity stayed hidden for longer,” the company said.

Security firm ESET has also seen scams and malware delivery leveraging DeepSeek’s newly gained notoriety. In one case, a fake DeepSeek website delivered digitally signed malware to users who clicked a ‘download now’ button.

Related: DeepSeek Compared to ChatGPT, Gemini in AI Jailbreak Test

Related: DeepSeek Security: System Prompt Jailbreak, Details Emerge on Cyberattacks

Related: Unprotected DeepSeek Database Exposed Chats, Other Sensitive Information

Related: What is DeepSeek, the Chinese AI Company Upending the Stock Market?

Related Content

Application Security

Dependabot gets a three-day cooldown window before opening pull requests, and PyPI rejects file uploads to releases older than 14 days.

Malware & Threats

The malware-as-a-service operation launches legitimate browsers on an invisible desktop, giving attackers persistent and covert remote access to compromised Windows systems.

Artificial Intelligence

SentinelOne’s new benchmark, built on the Fast16 case, shows which AI models can sustain a malware investigation and which cannot.

Malware & Threats

Part of a larger toolkit, HollowGraph uses a compromised 365 account’s calendar as a two-way dead-drop.

Malware & Threats

The zero-days CVE-2026-15409 and CVE-2026-15410 were exploited by a threat actor tracked by Volexity as UTA0533.

Malware & Threats

The new macOS malware has targeted at least 100 users to steal their passwords and cryptocurrency. 

Endpoint Security

Bitdefender researchers show how Windows bind links can create conflicting filesystem views to hide malware from endpoint security products.

Malware & Threats

The backdoor’s destructive capabilities include a standalone wiper, ransomware encryption, and a multi-pass wiping command.

Copyright © 2026 SecurityWeek ®, a Wired Business Media Publication. All Rights Reserved.

Exit mobile version