ZeroTrue
zerotrue.app
- Free plan
- Web
- iPhone and iPad
- Android
- API
- Self-hosted
What ZeroTrue is
ZeroTrue checks text, images, audio, video, and code for signs of AI generation through a web checker and API. It returns an overall probability with supporting details, such as highlighted regions or timestamps, for human review; results are evidence, not proof of who created something or why. The API follows one request pattern across the listed media types, with batch processing and asynchronous webhooks listed as capabilities. Python, TypeScript, and Go SDKs are available, alongside Android and iOS apps, a Telegram bot and Mini App, MCP connectivity, and an agent skill. The free plan costs nothing and includes 100 detections per month, API access, a web dashboard, and community support. For image checks on the free web tool, uploads must be JPG or PNG and no larger than 10 MB. Short samples, editing, compression, translation, re-recording, mixed human and AI content, and new generators can reduce confidence. The site identifies enterprise review, journalism and OSINT, platform integrity, and developer products as fit areas. Enterprise options list on-premise deployment, SSO and SAML, a custom SLA, and a dedicated account manager.
Who it suits
ZeroTrue may suit enterprise review teams, journalists and OSINT practitioners, platform integrity teams, and developers. Its tools cover several media types and include API access.
Good to know, and what to watch
- Checks text, images, audio, video, and code.
- Free plan includes API access.
- Lists Python, TypeScript, and Go SDKs.
- Supports multilingual detection.
- Free plan is limited to 100 detections per month.
- Free web image uploads are limited to 10 MB.
- Detection confidence can fall with edited or short samples.
Plans
Every plan and priceDeveloper
Enterprise
Growth
Free
Free plan · paid from $0.01
The verdict
ZeroTrue provides probabilistic signals across multiple content types, with web, API, and integration options. The site advises human review and says results should not be the sole basis for high-impact decisions.
What the maker publishes
- API
- The API uses one request pattern across text, images, audio, video, and code, with batch processing and asynchronous webhooks listed as capabilities. source
- API endpoint
- The documented check endpoint is POST /api/v1/check and accepts private-scan, deep-scan and idempotency options. source
- Async processing
- The API documentation describes polling and webhook delivery for long-running media checks. source
- Audio detection
- The audio detector examines voice-cloning and synthetic-speech signals including timing, spectral and speaker-consistency evidence. source
- Batch processing
- Batch detection is listed as an API capability for processing multiple files in one operation. source
- Beta status
- The terms describe the service as early beta and warn that features may be incomplete, unstable or changed without notice. source
- Code detection
- The code detector looks for patterns associated with output from coding assistants and large language models. source
- Company
- The site identifies the company as ZeroTrue Inc. source
- Company facts
- The official site footer identifies the operating entity as ZeroTrue Inc.; no headquarters or founding year is stated on the reviewed pages. source
- Company project date
- The Uberwow portfolio lists ZeroTrue as a 2025 multimodal AI-content and deepfake-detection project. source
- Data retention
- Terms say uploaded source files and submitted content are generally retained for no longer than 180 days, subject to stated exceptions. source
- Detection limitation
- ZeroTrue states that scores are probabilistic evidence and cannot prove authorship or intent. source
- Detection limits
- Short samples, editing, compression, translation, re-recording, mixed human and AI content, and new generators can reduce confidence. source
- Detection weakness
- Short samples, editing, compression, translation, re-recording, mixed authorship and newer generators can reduce confidence. source
- Documentation depth
- Public documentation includes API examples, response fields, methodology, model-evaluation guidance and modality-specific guides. source
- Enterprise deployment
- The Enterprise plan lists on-premise deployment, SSO and SAML, a custom SLA, and a dedicated account manager. source
- Evidence
- Results provide probabilistic evidence and supporting signals, not proof of authorship or intent. source
- Explainable results
- Results can include probability scores, confidence bands, highlighted regions, timestamps, forensic indicators and provenance notes. source
- Image detection
- The image detector checks diffusion artifacts, GAN output, face swaps, metadata and other manipulation indicators. source
- Image upload limit
- The free web image checker accepts JPG and PNG files up to 10 MB. source
- Independent description
- Parse describes ZeroTrue as a multimodal detector with explainable confidence scores and a unified API for enterprise privacy and security workflows. source
- Independent visibility metric
- Parse assigns ZeroTrue a Parse Score of 47.3 based on its AI-answer tracking dataset. source
- Integrations
- The site describes Android and iOS apps, a Telegram bot and Mini App, MCP connectivity, an agent skill, and SDKs. source
- Integrations
- The product site advertises Android and iOS apps, a Telegram bot and Mini App, MCP connectivity, an agent skill and SDK integrations. source
Checked against the maker's pages on 21 Oct 2026.