TL;DR
A LinkedIn user embedded a prompt injection in their profile, making AI-driven recruiter responses appear in Old English. This demonstrates how AI profiles can be manipulated, raising concerns about AI security and spam.
A LinkedIn user inserted a prompt injection into their profile, causing AI-based recruiters to respond in Old English, exposing potential vulnerabilities in AI-driven recruitment processes.
On May 15, 2026, a software developer known as tmuxvim shared on social media that they had embedded a prompt injection into their LinkedIn bio. This prompt instructed AI scanning their profile to address them as ‘My Lord’ and speak exclusively in Old English, appropriate for the year 900 AD. As a result, AI recruiters attempting to contact them responded with messages written in Old English, including a lengthy message from an AI claiming to be from a venture capital firm, filled with archaic language and references to treasure hoards.
The AI-generated message began with ‘My Lord Arthur’ and included a paragraph in Old English about a hoard of gold and a warrior’s craft, which was largely unintelligible but demonstrated the injection’s effect. The incident was confirmed by tmuxvim, who shared screenshots and the text of the responses. The demonstration underscores how AI tools used in recruitment can be manipulated through prompt injections embedded in user profiles.
Why It Matters
This incident highlights a potential security vulnerability in AI-powered recruitment systems, where prompt injections can cause AI agents to produce nonsensical or humorous responses. It raises concerns about the reliability of AI in professional settings and the need for safeguards against manipulation. As AI becomes more integrated into hiring processes, understanding and mitigating such vulnerabilities is increasingly important for companies and users alike.
Secure AI Agents with LangChain, MCP, and Tool-Using LLMs: A Developer’s Guide to Safe Invocation,Prompt Defense, and Context-Aware Generative Workflows
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Background
Prompt injection attacks have been a known issue in AI systems, where malicious or unintended prompts alter AI behavior. This specific case involves a user embedding a prompt in their LinkedIn bio to influence AI responses. The incident follows broader discussions about AI security and trustworthiness, especially as AI tools are adopted more widely in recruitment and professional networking. Prior to this, similar vulnerabilities have been observed in chatbots and other AI applications, but this is among the first publicly documented cases involving a professional social media profile.
“I put a prompt injection into my LinkedIn bio and recruiters are messaging me in Old English and calling me Lord.”
— tmuxvim
“This demonstrates how easily AI systems can be manipulated through user profiles, raising questions about the security of AI-driven recruitment tools.”
— tech security analyst
AI projects with Raspberry Pi: High-performance artificial intelligence for robotics, security, home automation, and vision (Essentials)
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
What Remains Unclear
It is not yet clear how widespread this vulnerability is across different AI recruitment tools or whether platforms will implement safeguards to prevent such prompt injections. The long-term impact on AI trust and security remains to be seen.
AI profile verification tools
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
What’s Next
Recruitment platforms and AI developers are likely to review and update their security measures to prevent prompt injections. Further incidents or demonstrations may emerge, prompting industry discussions on AI safety protocols and profile verification processes.
OpenClaw AI Agent on Arduino UNO Q: Smart Home Automation & Security Projects for Beginners and Enthusiasts
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Key Questions
Can AI responses be reliably controlled or secured?
Current research and development are focusing on improving AI robustness, but vulnerabilities like prompt injections still pose risks. Users and developers need to be aware of potential manipulation and implement safeguards.
Does this mean AI in recruitment is unsafe?
This incident highlights a vulnerability but does not imply all AI recruitment tools are insecure. It underscores the importance of security measures and ongoing monitoring to prevent manipulation.
Will platforms ban prompt injections in profiles?
Platforms may introduce stricter profile verification or prompt filtering, but specific measures are not yet confirmed. Industry response will shape future policies.
Could this kind of manipulation impact hiring decisions?
While unlikely to directly influence hiring decisions, such manipulations could undermine trust in AI systems and lead to false or misleading interactions.
