url: https://uncovai.com/faq/
dateModified: 2026-02-24T15:10:13Z
headline: FAQ - UncovAI
description: Browse the UncovAI FAQ to find quick answers to common questions about AI content detection, accounts, usage, and support.
text: FAQs FAQs: Everything you need to know about UncovAIand our detectors What is UncovAI? UncovAI is an AI content detection platform that identifies whether text, images, video, audio, and URLs were created by a human or generated by artificial intelligence. Built in France and backed by NVIDIA Inception and Microsoft for Startups, UncovAI serves businesses, governments, and individuals who need reliable verification of digital content authenticity. Its detection engine is built on proprietary mathematical models rather than heavy Deep Learning systems, making it both accurate and energy-efficient. What makes UncovAI different from other AI detectors? Most AI detection tools focus on a single media type — typically text or images. UncovAI covers all five: text, images, video, audio, and phishing URLs from one platform. Real-time meeting detection – UncovAI detects deepfake voices and synthetic video during active Zoom, Microsoft Teams as as they happen — not after the fact. Works where you already are – Available as a web app, browser extension, WhatsApp bot, meeting integration, and enterprise API. Energy-efficient by design – Built on mathematical models rather than power-intensive Deep Learning systems, delivering strong accuracy with a significantly lower carbon footprint. Zero data retention – No content submitted for analysis is ever stored, shared, or used beyond the immediate detection request. Institutional trust – Selected by NVIDIA Inception and Microsoft for Startups programs, and deployed by enterprise and government clients across Europe and the World. What types of AI-generated content can UncovAI detect? AI Text Detection — Identifies content written by ChatGPT, Claude, Gemini, LLaMA, Mistral, Copilot, and services built on these models. Used by educators, HR teams, journalists, and compliance professionals. AI Image Detection — Detects synthetic images from DALL-E, Midjourney, Stable Diffusion, Leonardo AI, Ideogram, Adobe Firefly, and Canva AI. Identifies manipulated faces, generated scenes, and synthetic visuals used in fraud and disinformation. AI Video Detection — Analyzes video for deepfake manipulation, face-swapping, and synthetic generation. Relevant for KYC verification, media authentication, and legal evidence validation. AI Audio & Voice Clone Detection — Detects AI-synthesized speech produced by ElevenLabs, Murf, Resemble AI, and similar platforms. Protects against voice impersonation, phone scams, and identity fraud. Phishing URL Detection — Identifies AI-generated phishing links and malicious websites before users interact with them, reducing exposure to social engineering attacks. Can UncovAI detect deepfakes during live video calls? Yes. UncovAI integrates with Zoom, Microsoft Teams, and Google Meet to analyze audio and video streams in real time. It identifies AI-generated voices, voice clones, and deepfake video during active calls, alerting users to potential impersonation before any damage occurs. This makes UncovAI one of the very few detection tools that operates at the moment of communication rather than retrospectively. What is CEO fraud and how does UncovAI help prevent it? CEO fraud is a social engineering attack where criminals impersonate a senior executive — using AI-generated voice clones or deepfake video — to instruct employees to transfer funds, share credentials, or take damaging actions. The FBI estimates Business Email Compromise and related schemes have cost organizations over $50 billion globally. UncovAI helps prevent CEO fraud by detecting AI-generated voices in real time during calls, identifying deepfake video in meeting contexts, and flagging AI-written phishing emails before they reach decision-makers. Which models can UncovAI detect? UncovAI is model-agnostic and regularly updated as new generative systems emerge. Text: ChatGPT (GPT-3.5, GPT-4, GPT-4o), Claude, Gemini, LLaMA, Mistral, Copilot, and third-party services built on these APIs. Images: DALL-E 2 & 3, Midjourney, Stable Diffusion, Leonardo AI, Ideogram, Adobe Firefly, Canva AI, and Bing Image Creator. Audio & Voice: ElevenLabs, Murf, Resemble AI, Descript Overdub, and other neural voice synthesis platforms. Detection coverage is continuously expanded to keep pace with the evolving generative AI landscape. Is UncovAI private and secure? Yes. UncovAI operates on a strict zero data retention policy. No content submitted for analysis is stored, logged, shared, or used for any purpose beyond the immediate detection request. For organizations with heightened data sovereignty requirements, UncovAI offers fully on-premises deployment, ensuring sensitive content never leaves your own infrastructure. Who uses UncovAI? Businesses & Enterprises — Finance teams detecting CEO fraud, HR departments verifying candidate authenticity, compliance officers checking document integrity, and security teams neutralizing AI-driven phishing. Government & Public Institutions — Security agencies identifying AI-manipulated media, public bodies protecting institutional communications, and teams monitoring AI-generated disinformation. Media & Journalism — Reporters and editors verifying images, videos, and sources before publication. Fact-checkers identifying synthetic content at scale. Education — Academic institutions detecting AI-written submissions and maintaining integrity standards in the era of generative AI. Legal & Insurance — Professionals verifying the authenticity of digital evidence and AI-manipulated claims documentation. Individuals — Anyone needing to verify a suspicious voice message, image, or video received via WhatsApp, email, or social media Does UncovAI support multiple languages? UncovAI performs with highest accuracy in English and French. Early validation studies show promising detection performance in Ukrainian. The team is actively expanding language support and welcomes user feedback on detection performance across other languages. How much does UncovAI cost? UncovAI offers a freemium model allowing users to test core detection features at no cost. Paid plans provide increased detection volume, access to all media types, and advanced features. Enterprise pricing covers API access, custom integration, SLA agreements, and on-premises deployment. Full plan details are available at uncovai.com/pricing. Organizations with specific requirements can contact the team directly for a tailored proposal. Is UncovAI compliant with GDPR and EU AI Act requirements? UncovAI is built and headquartered in France, operating within the European Union’s legal framework. Its zero data retention policy and privacy-first architecture are aligned with GDPR requirements. As the EU AI Act comes into force — particularly provisions around synthetic media disclosure and deepfake regulation — UncovAI’s detection capabilities are directly relevant to organizations building AI compliance programs across Europe. Why does AI content detection matter? Generative AI has fundamentally changed how digital content is created and distributed. Voice cloning can replicate a person’s voice from a short audio sample. Deepfake video is increasingly difficult to distinguish from genuine footage. AI-written phishing emails have demonstrated higher engagement rates than human-written equivalents. In this environment, the ability to verify whether content is authentic is becoming a baseline requirement for trust, security, and informed decision-making across business, government, and everyday life. UncovAI provides the tools to make that verification fast, accessible, and private. How accurate is UncovAI's detection? UncovAI delivers forensic-level detection validated across large volumes of content samples. Text detection results are returned on a five-point confidence scale: 🤖 Most likely generated by AI 🤖 Likely generated by AI 😐 Difficult to determine 🧠 Likely written by a human 🧠 Most likely written by a human Image detection returns: Likely Generated or Likely Human, supported by underlying confidence scoring. UncovAI recommends treating results as strong forensic indicators to inform — not replace — professional judgment. Because generative AI evolves rapidly, no detection system can guarantee absolute certainty in every case. Images processed by smartphone AI enhancement features may occasionally return AI-generated signals. Where can I use UncovAI? Web App — Full-featured detection for text, images, video, audio, and URLs via a secure browser interface. Chrome Browser Extension — Instant authenticity checks on articles and social media content as you browse. WhatsApp Bot — Send any suspicious message, voice note, image, or video to UncovAI via WhatsApp for immediate analysis. Meeting App Integration — Real-time deepfake and voice clone detection during live Zoom, Teams, and Google Meet sessions. Enterprise API — Integrate UncovAI’s detection engine into your own applications or security workflows via REST API. On-Premises Deployment — Full local installation for organizations requiring complete data sovereignty. Can’t find an answer? Let’s talk! feel free to get in touch with us if you’re intrigued by our technologyor if you’re interested in exploring potential partnerships! Contact us Are you sure you want to proceed with the payment? Confirm Cancel
image: