Customize Consent Preferences

We use cookies to help you navigate efficiently and perform certain functions. You will find detailed information about all cookies under each consent category below.

The cookies that are categorized as "Necessary" are stored on your browser as they are essential for enabling the basic functionalities of the site. ... 

Always Active

Necessary cookies are required to enable the basic features of this site, such as providing secure log-in or adjusting your consent preferences. These cookies do not store any personally identifiable data.

No cookies to display.

Functional cookies help perform certain functionalities like sharing the content of the website on social media platforms, collecting feedback, and other third-party features.

No cookies to display.

Analytical cookies are used to understand how visitors interact with the website. These cookies help provide information on metrics such as the number of visitors, bounce rate, traffic source, etc.

No cookies to display.

Performance cookies are used to understand and analyze the key performance indexes of the website which helps in delivering a better user experience for the visitors.

No cookies to display.

Advertisement cookies are used to provide visitors with customized advertisements based on the pages you visited previously and to analyze the effectiveness of the ad campaigns.

No cookies to display.

shanghaishanghai
0

The AI world is buzzing about Manus, a new general-purpose AI Agent that has taken the internet by storm. Boasting the ability to handle complex file processing, data analysis, and code writing, Manus quickly became the must-have AI tool, with users scrambling for coveted invitation codes. But the hype train may have hit a snag.

Just days after its launch, Manus appears to have been compromised. A user named jian on X (formerly Twitter) claims to have successfully extracted crucial system information and runtime code from the AI agent with a single, simple command.

Instead of waiting for an invitation, users can now explore open-source alternatives. The MetaGPT team has developed OpenManus, a free, open-source replica that doesn’t require an invitation code and can be run locally. CAMEL-AI has also released OWL, another open-source alternative to Manus.

The Hack: A Simple Request, a Revealing Response

According to reports, jian simply asked Manus to output the files located in the /opt/.manus/ directory. To the surprise of many, Manus complied, revealing potentially sensitive information about its internal workings and sandboxed runtime environment.

[Manus link: https://manus.im/share/lLR5uWIR5Im3k9FCktVu0k?replay=1]

[Image of Manus outputting information]

The implications of this alleged breach are significant. The exposed information could provide insights into the system prompts guiding Manus’s behavior, the architecture of the underlying large language model (LLM) powering the agent, and potential vulnerabilities that could be exploited.

Community Reacts, CTO Responds

News of the alleged hack has spread rapidly throughout the AI community, sparking discussions about the security of AI agents and the importance of robust security measures.

While details remain scarce, the incident raises questions about the safeguards in place to protect proprietary information and prevent unauthorized access to sensitive system components.

As of press time, the CTO of Manus has responded to the claims, but details of the response are still emerging. This is a developing story, and we will continue to update this article as more information becomes available.

The Future of AI Agent Security

This incident serves as a stark reminder of the challenges and risks associated with the rapid development and deployment of AI agents. As these technologies become increasingly sophisticated and integrated into critical systems, ensuring their security and resilience will be paramount.

The AI community must prioritize the development of robust security protocols, ethical guidelines, and transparency measures to mitigate potential risks and build trust in these powerful tools. The future of AI depends on it.

References:

  • Machine Heart News Report: [Original Article Link] (Replace with the actual link when available)
  • X (Twitter) Post by jian: [Link to jian’s Tweet] (Replace with the actual link when available)
  • MetaGPT’s OpenManus: [Link to OpenManus Repository] (Replace with the actual link when available)
  • CAMEL-AI’s OWL: [Link to OWL Repository] (Replace with the actual link when available)

Note: This article is based on preliminary reports and information available at the time of writing. Further investigation and verification are needed to confirm the details of the alleged hack and its potential impact.


>>> Read more <<<

Views: 0

0

发表回复

您的邮箱地址不会被公开。 必填项已用 * 标注