Frequently Asked Questions

TextGuardAI is an all-in-one platform to verify the authenticity of any content. You can detect AI-generated text, images, video and audio, and you can humanize AI-sounding text so it reads as if a person wrote it — all from a single account, with no software to install.

There is no permanent free tier, but you can unlock the entire platform with a 24-hour Trial for just €0.70. The Trial gives you full access to text, image, video and audio detection plus the Humanizer, so you can try everything before committing to a monthly plan.

For a one-time €0.70 you get 24 hours of full access with sampler quotas (1 image, a few analyses, 30 seconds of video and 60 seconds of audio). After 24 hours the Trial automatically converts to the Gold plan unless you cancel before it ends — and you can cancel in one click from your dashboard at any moment. The Trial can be used once per account.

No. TextGuardAI runs entirely in your browser on any device — desktop, tablet or phone. Just sign in and start analyzing or humanizing text right away.

Students and teachers checking academic integrity, writers and editors verifying originality, recruiters reviewing applications, content and marketing teams protecting brand authenticity, and anyone who needs to know whether a piece of content was made by a human or by AI.

Paste your text and TextGuardAI returns an AI-probability score from 0 to 100, plus a clear verdict: AI Generated, Mixed Content or Human Written. You also get a confidence level (high, medium or low) so you know how sure the result is. Detection is powered by GPTZero, one of the most established detection engines available.

Instead of a single number, you see every sentence scored individually and color-coded: green for human, red for AI, and a lighter red for mixed. You can filter the view to show only the AI, mixed, human or flagged sentences, which makes it easy to pinpoint exactly which parts of a document look machine-written.

It uses advanced statistical language models that achieve high accuracy on typical texts, and it also reports a confidence level and a 'burstiness' signal to back up each result. That said, no AI detector on the market is infallible — very short, heavily edited or translated texts are harder to classify, so we recommend using the result as strong evidence rather than absolute proof.

Text detection is optimized for English, Spanish and French. You can paste text in those languages and get reliable sentence-level analysis. (Note: the app interface itself is available in 30 languages — that's separate from the languages the detector analyzes.)

Yes. When AI is detected, TextGuardAI can identify the likely type: purely AI-generated, AI text that was paraphrased or run through a 'bypasser' tool, or human text that was concatenated or polished with AI. This nuance helps you understand exactly what you're looking at.

Each analysis accepts up to 15,000 characters, with a minimum of 5 words. For longer documents you simply split them into sections. Your monthly word allowance depends on your plan (100,000 words on Basic, 1,000,000 on Gold).

Yes. The engine is built to recognize the statistical patterns of modern large language models, including the most popular AI writing tools. Because new models appear constantly, we treat the score as a strong probability signal rather than a guarantee for any single tool.

Yes. Upload an image and TextGuardAI tells you whether it was AI-generated and whether it's a deepfake. When possible it even identifies the likely generator family — diffusion models like Midjourney, DALL·E and Stable Diffusion, or GAN-based tools.

JPEG, PNG, WebP, GIF, HEIC and HEIF, up to 10 MB per image. Images are processed securely and never shared.

Yes. Video is analyzed frame by frame for both AI generation and deepfake manipulation, and you get a per-frame breakdown plus an overall verdict. A separate deepfake score appears whenever a face is detected in the footage.

MP4, WebM and MOV, up to 500 MB per file. The maximum duration per video depends on your plan: 30 seconds on Trial, 1 minute on Basic and 2 minutes on Gold, with a monthly pool of video minutes on each paid plan.

Yes. Upload an audio file and TextGuardAI analyzes it segment by segment to flag synthetic or AI-generated speech, returning a confidence score for each part of the recording.

Common formats including MP3, WAV, FLAC, OGG, AAC, M4A and more, up to 50 MB. Maximum duration per file is 60 seconds on Trial, 2 minutes on Basic and 5 minutes on Gold, with a monthly audio-minutes pool on the paid plans.

It rewrites robotic, AI-sounding text so it reads as if a thoughtful person wrote it — varying sentence rhythm, removing tell-tale AI vocabulary and adding natural human phrasing. It's perfect for turning a rough AI draft into text that feels authentic and original.

Yes. The Humanizer is built on a strict fidelity rule: it never invents facts, keeps every idea from your original, and stays within about ±20% of the original length. You get a natural rewrite, not a different message.

Three: Standard (a versatile, well-crafted voice), Casual (warm and conversational, great for blogs and social posts) and Academic (precise terminology and measured phrasing for essays and papers). Pick the one that matches your audience.

The Humanizer is specifically engineered to reduce the signals that detectors look for — predictable word choices, uniform sentence length and typical 'AI vocabulary'. It significantly lowers AI-detection scores in most cases, but because every detector is different and constantly updated, we can't promise it will bypass any specific tool 100% of the time.

Any language. The Humanizer automatically detects the language of your text and rewrites it in that same language, so you can humanize content in English, Spanish, French, German and far beyond.

Between 10 and 5,000 words per request, and the rewrite streams in word by word so you see it appear in real time. Your monthly word allowance depends on your plan.

Three: the 24-hour Trial (€0.70, a full sampler of every tool), Basic (€29.90/month) and Gold (€39.90/month, our most popular plan). All prices are shown in your local currency.

Basic includes 100,000 words/month, 50 images, 5 minutes of video and 10 minutes of audio. Gold gives you 10× the words (1,000,000/month), 200 images, 30 minutes of video, 60 minutes of audio, longer per-file limits and priority support. For just a little more per month, Gold is the best value for regular users.

Start with the €0.70 Trial to explore everything. If you analyze content occasionally, Basic is plenty. If you work with AI content daily — long documents, lots of images, or video and audio — Gold's much larger allowances and priority support pay for themselves quickly.

Each text detection, image, video, audio check or humanize counts as one operation against your plan. Paid plans (Basic and Gold) don't cap the number of analyses — you're only limited by your monthly words, images and media minutes.

You'll see a clear in-app message telling you which allowance is used up (for example words or video minutes). Your other allowances keep working, and everything resets at your next monthly renewal. You can also upgrade to Gold instantly for more headroom.

Yes. On Basic and Gold all allowances refresh at the start of each monthly billing cycle, so unused quota doesn't carry over but you get a full new allowance every month.

The Trial is a one-time €0.70 for 24 hours. Basic is €29.90/month and Gold is €39.90/month. Prices are shown in your local currency automatically.

14 currencies, including Euro, US Dollar, British Pound, Canadian and Australian Dollar, Swiss Franc, Japanese Yen, Korean Won, Brazilian Real, Mexican Peso, Singapore Dollar and Gulf currencies. We detect your country automatically and show the right currency — no manual conversion needed.

Yes, in one click from your dashboard, with no penalty. You keep full access until the end of the period you already paid for, and you can reactivate just as easily if you change your mind.

Because access is granted instantly, we don't issue refunds for partial use of a period that has already been billed. You can always cancel to prevent the next renewal — you'll simply keep access until that period ends.

All major credit and debit cards, plus one-click options like Link, Apple Pay and Google Pay where available. Payments are processed securely by Stripe.

Completely. Card details are handled directly by Stripe on its PCI-DSS Level 1 certified infrastructure. TextGuardAI never sees or stores your full card number, CVV or expiry.

Yes, Basic and Gold renew automatically each month so your service never gets interrupted. You can turn off renewal anytime by cancelling — access continues until the current period ends.

New users can unlock a 20% discount on their first subscription. The discount applies to your first charge — an easy way to start with Gold for less.

Your analyses are saved to your private history so you can revisit and re-open them anytime — this is part of the service. Everything is encrypted in transit and at rest, stored on servers within the European Economic Area, and isolated so that only you can access your own content. You can delete any analysis, or wipe your entire history, whenever you want.

On secure servers inside the EEA, encrypted with TLS 1.3 in transit and encrypted at rest. Each account is isolated at the database level, so no other user can ever see your content.

Yes. You can delete individual analyses (with a short undo window in case of mistakes) or bulk-delete your whole history in one confirmed action. Deleted items are fully purged after a short safety period. You also have full GDPR rights, including erasure on request.

Yes. TextGuardAI is operated by BITOR SOFTWARE SL, a company based in Barcelona, Spain, and fully complies with the GDPR. You can exercise your rights of access, rectification, erasure, portability and objection at any time by contacting support@textguardai.com.

To deliver results we send your content, encrypted, to specialized processors: GPTZero for text detection and DeepSeek for humanization, alongside Supabase (secure EU hosting) and Stripe (payments). They act strictly as data processors and never use your content for their own purposes.

Yes. To protect your privacy and keep your history secure, all tools require an account. Creating one takes seconds, and the €0.70 Trial unlocks everything immediately.

Yes. Your dashboard keeps a full history of every detection and humanization, which you can filter, reopen and even replay — re-opening a past result costs nothing against your quota. Live usage bars also show exactly how much of each allowance you've used.

The interface is available in 30 languages, including English, Spanish, French, German, Portuguese, Italian, Dutch, Polish, Arabic, Chinese, Japanese, Korean, Thai, Vietnamese and many more. Switch anytime from your settings.

Everything is in your dashboard: see your plan and usage, upgrade or switch plans, update your payment method through the secure Stripe portal, download invoices, and cancel or reactivate in one click.