OP 22 May, 2025 - 12:58 AM
A new AI chatbot called Venice.ai has gained popularity in underground hacking forums due to its lack of content restrictions.
According to a recent investigation by Certo, the platform offers subscribers uncensored access to advanced language models for just $18 a month, significantly undercutting other dark web AI tools like WormGPT and FraudGPT, which typically sell for hundreds or even thousands of dollars.
What sets Venice.ai apart is its minimal oversight. The platform stores chat histories only in users’ browsers, not on external servers, and markets itself as “private and permissionless.”
This privacy-focused design, combined with the ability to disable remaining safety filters, is reportedly proving especially attractive to cybercriminals.
Unlike mainstream tools such as ChatGPT, Venice.ai can reportedly generate phishing emails, malware and spyware code on demand.
In testing, Certo said it successfully prompted the chatbot to create realistic scam messages and fully functional ransomware. It even generated an Android spyware app capable of recording audio without user knowledge – behavior that most AI platforms would reject outright.
Advanced Threat Capabilities with Minimal Effort
Certo’s findings suggest that Venice.ai goes further than simply ignoring harmful queries. It appears to have been configured to override ethical constraints altogether.
In one example, it reasoned through an illegal prompt, acknowledged its malicious nature and proceeded anyway.
The generated output included:
- C# keyloggers designed for stealth
- Python-based ransomware with file encryption and ransom notes
- Android spyware complete with boot-time activation and audio uploads
Certo’s report highlights a growing challenge: as AI tools become more powerful and easier to access, so does their potential for misuse.
Source:
InfoSecurity
https://www.infosecurity-magazine.com/ne...rsecurity/
According to a recent investigation by Certo, the platform offers subscribers uncensored access to advanced language models for just $18 a month, significantly undercutting other dark web AI tools like WormGPT and FraudGPT, which typically sell for hundreds or even thousands of dollars.
What sets Venice.ai apart is its minimal oversight. The platform stores chat histories only in users’ browsers, not on external servers, and markets itself as “private and permissionless.”
This privacy-focused design, combined with the ability to disable remaining safety filters, is reportedly proving especially attractive to cybercriminals.
Unlike mainstream tools such as ChatGPT, Venice.ai can reportedly generate phishing emails, malware and spyware code on demand.
In testing, Certo said it successfully prompted the chatbot to create realistic scam messages and fully functional ransomware. It even generated an Android spyware app capable of recording audio without user knowledge – behavior that most AI platforms would reject outright.
Advanced Threat Capabilities with Minimal Effort
Certo’s findings suggest that Venice.ai goes further than simply ignoring harmful queries. It appears to have been configured to override ethical constraints altogether.
In one example, it reasoned through an illegal prompt, acknowledged its malicious nature and proceeded anyway.
The generated output included:
- C# keyloggers designed for stealth
- Python-based ransomware with file encryption and ransom notes
- Android spyware complete with boot-time activation and audio uploads
Certo’s report highlights a growing challenge: as AI tools become more powerful and easier to access, so does their potential for misuse.
Source:
InfoSecurity
https://www.infosecurity-magazine.com/ne...rsecurity/
