Product Introduction
- Truely by Validia is a real-time AI assistant detection tool designed to identify candidates using artificial intelligence during live interviews. The solution requires candidates to download a lightweight desktop application that monitors system activity and communication patterns during virtual interviews. It operates transparently in the background while maintaining compatibility with standard video conferencing platforms like Zoom or Microsoft Teams.
- The core value lies in preserving interview integrity by detecting unauthorized AI assistance during high-stakes hiring processes. It provides hiring teams with actionable insights to differentiate between authentic human responses and AI-generated content. This ensures organizations evaluate candidates based on genuine skills rather than automated support systems.
Main Features
- The tool performs continuous audio/video analysis using proprietary algorithms to detect anomalies in response latency, vocal patterns, and linguistic complexity indicative of AI tools like ChatGPT. It flags suspicious activity through a dashboard updated every 3-5 seconds during interviews.
- A sub-100MB native application requires no administrative privileges or complex setup, minimizing friction for candidates. The app automatically terminates post-interview and generates encrypted activity logs for audit purposes.
- Real-time compatibility with macOS 10.15+ ensures functionality across modern Apple devices commonly used in professional settings. Future Windows support is planned for Q4 2025 based on roadmap documents.
Problems Solved
- Organizations lose an estimated $23,000 per mis-hire due to undetected AI-assisted interviews according to Validia's internal research. Traditional proctoring solutions fail to detect AI tools operating through secondary devices or browser extensions.
- The primary users are technical hiring managers, HR departments at enterprises, and government agencies conducting security-sensitive recruitment. Secondary users include educational institutions verifying oral exam authenticity.
- Typical scenarios include detecting candidates using covert AI speech-to-text-to-speech pipelines during coding interviews, identifying real-time translation abuse in language proficiency tests, and flagging scripted responses in consulting case interviews.
Unique Advantages
- Unlike screen recording tools like HireVue, Truely analyzes low-level system processes and network traffic to detect AI tools running locally or through cloud services. This includes identification of LLM APIs and hidden virtual machines.
- The patent-pending "Digital Vocal Print" technology distinguishes between human voice modulation and AI-generated speech patterns by analyzing 120+ acoustic parameters at 20ms intervals.
- Competitive edge stems from zero latency in detection alerts compared to post-interview analysis tools. Validia's database contains behavioral fingerprints of 47 known interview AI tools updated hourly through threat intelligence feeds.
Frequently Asked Questions (FAQ)
- How does Truely differentiate between legitimate research and AI cheating? The tool ignores predefined allowlists of approved websites/documentation portals while flagging interactions with AI writing assistants, code generators, and paraphrasing tools through process recognition and network signature analysis.
- What system requirements apply for candidates? Currently requires macOS 10.15 or newer with 8GB RAM and microphone access. The app automatically blocks installation on virtual machines or devices with detected debugging tools.
- How is candidate privacy maintained? Data is retained for 30 days in AES-256 encrypted format, with GDPR-compliant opt-out mechanisms. The app cannot access personal files, emails, or non-interview related applications.