ZeroTrue

ZeroTrue is a detector for AI-generated text, images, audio, video, and code, with web and API access. It suits teams that need to triage several media types and can have people review the signals; it is not a sound choice when a score must settle authorship or intent.

Overview

Its central advantage is the range of formats handled by one service. Results pair an overall probability with supporting details, such as highlighted regions or timestamps, giving reviewers leads to investigate rather than a definitive answer. That distinction matters: detection is probabilistic evidence, not proof.

Key features

Signals and limitations

Review cues can help focus attention, but confidence may fall with short samples, editing, compression, translation, re-recording, mixed human and AI content, or newer generators. Multilingual coverage helps broaden screening, not remove those constraints. ZeroTrue cautions against using detector output alone for high-impact decisions, so organizations need a human review step.

API and integrations

One API request pattern covers all five media types, avoiding separate modality-specific interfaces. Batch processing and asynchronous webhooks suit workflows handling groups of items or results that need not return immediately. Python, TypeScript, and Go SDKs add developer options; Android and iOS apps, a Telegram bot and Mini App, MCP connectivity, and an agent skill provide additional access points.

Security and deployment

The public website and application endpoints use HTTPS, but that does not establish whether a particular deployment meets an organization’s data requirements. Confirm data location, retention, deletion, access controls, and deployment needs before production use. Enterprise offers on-premise deployment, SSO and SAML, a custom SLA, and a dedicated account manager for organizations needing those arrangements.

The free web image checker accepts JPG and PNG files up to 10 MB, a useful but bounded route for occasional image checks.

Pricing

The Free plan costs nothing and includes 100 detections per month, API access, a web dashboard, and community support. That is a meaningful starting allowance for evaluation or light use, but the monthly cap makes it unsuitable for sustained high-volume screening.

Developer is billed pay-as-you-go per modality, with unlimited detections, Priority Support, and rates of $0.01 / image, $0.05 / audio min, and $0.10 / video min. The per-unit rates make costs easier to relate to usage, though audio and video volume can accumulate; no price is given for text or code. Growth uses custom pricing billed by volume, adding volume discounts, batch processing API, asynchronous webhooks, and a dedicated Slack channel. It is a better fit for substantial throughput than the capped Free plan, but requires custom pricing.

Enterprise is billed custom and adds SSO and SAML, on-premise deployment, a custom SLA, and a dedicated account manager. Those are organizational controls and support rather than a necessary upgrade for every individual reviewer. Growth and Enterprise have no published fixed price, so buyers need a quote to compare their costs.

Platforms

ZeroTrue supports Android, iOS, web, API, and self-hosted use. That mix can accommodate both direct checking and integration into products or organizational deployments, while the security and data questions still need to be resolved for production.

Who it's for

Journalists and OSINT researchers can use the signals to prioritize material for verification; platform-integrity teams can screen across varied media; and developers can build checks into products through the shared API and SDKs. It is a poor fit for anyone seeking an authorship verdict without human assessment, or for organizations unwilling to resolve data-handling requirements before deployment.

Pros and cons

• Broad media coverage: one service checks text, images, audio, video, and code, useful for mixed-format review.
• Actionable review cues: probabilities plus highlighted regions or timestamps offer more direction than a bare score, while remaining inconclusive.
• Developer options: a common API pattern, SDKs, batch processing, and webhooks support product and larger-scale workflows.
• Material confidence limits: common transformations and mixed-origin content can weaken results, requiring human review.
• Free usage is capped: 100 monthly detections may be too little for teams with regular screening needs.
• Production diligence remains necessary: HTTPS does not answer data location, retention, deletion, access, or deployment questions.

Alternatives

Choose Trinka if its broader platform range or freemium option better matches your workflow. Grammarly is an alternative with a free plan and an allowance of 100 AI prompts per month on its Free plan. Sapling offers a free plan with 2,000 AI detector characters per check, five detector files per batch, and 20 snippets, which may suit bounded text checks.

SlopProx is another alternative. AI Detect may suit users seeking a free, local-use option with no usage limits and no API key. AI Detector Pro offers a free plan with three AI scans per month. Detecto is another freemium alternative. GPTZero offers a free plan with 10,000 words per month and 50,000 characters, making it worth considering for a text-focused allowance.

Verdict

ZeroTrue is a strong fit for teams that need to screen multiple media types through one API or a range of apps and can treat results as prompts for investigation. Its breadth and review signals are the main reasons to choose it; confidence limitations and unresolved production data requirements are reasons to look elsewhere when a conclusive, low-friction decision process is essential.