Question 1

Is there really a free plan?

Accepted Answer

Yes. The free tier includes 10 scans per day on text and image verification. No credit card required.

Question 2

What happens if I hit my daily limit?

Accepted Answer

You can wait for the next day or upgrade to Pro instantly. Your scan count resets every 24 hours.

Question 3

Can I cancel anytime?

Accepted Answer

Yes. Monthly plans cancel anytime with no hidden fees. You keep Pro access until the end of your billing period.

Question 4

Will the free tier stay?

Accepted Answer

Yes. We believe everyone deserves access to basic AI verification. The free tier is permanent.

Question 5

Why do you charge for Pro and Team?

Accepted Answer

Running ensemble AI models across text, image, audio, and video requires significant GPU compute. Paid plans help us improve accuracy, add new modalities, and keep the service running — without selling your data or showing ads.

Question 6

How accurate is Aiscern?

Accepted Answer

Our latest benchmarks show approximately 94% accuracy on text, 98% on images (14-layer ensemble including physics-based signals), 92% on audio, and 90% on video. Accuracy varies by content type, generator, and compression level. These are measured on public held-out test sets, not cherry-picked results.

Question 7

Can I use Aiscern for legal or academic decisions?

Accepted Answer

No. Verification results are probabilistic, not definitive. Always use human judgment for high-stakes decisions. Never use a single result as sole evidence in legal proceedings or academic integrity cases.

Question 8

What AI generators can you verify?

Accepted Answer

Our models are updated quarterly. We verify content from major generators including ChatGPT, GPT-4, Claude, Midjourney, DALL-E, Stable Diffusion, ElevenLabs, and common TTS tools. Novel generators released after our last update may evade verification until the next fine-tune.

Question 9

How does the ensemble work?

Accepted Answer

We run content through multiple independent verification signals and combine them into a single confidence score. No single model makes the final call — results are cross-verified before delivering a verdict.

Question 10

What does an "Uncertain" verdict mean?

Accepted Answer

An "Uncertain" result means the ensemble did not reach ≥62% confidence to label AI, nor ≤38% to label human. This is not a failure — it means the content is genuinely ambiguous. Try running a longer sample or checking a different modality.

Question 11

Does Aiscern work on languages other than English?

Accepted Answer

Text verification was primarily trained on English-language data. Non-English text may produce higher false-positive rates. Treat non-English results with extra caution. Multilingual support is on the roadmap.

Question 12

What happens to my uploads?

Accepted Answer

Files are processed for verification and deleted within 24 hours. We do not train our models on your content without explicit opt-in. Scan results are stored in your history for 12 months.

Question 13

Do you sell my data?

Accepted Answer

No. We do not sell, share, or use your submitted content for any purpose other than providing the verification service to you.

Question 14

Who built Aiscern?

Accepted Answer

Aiscern is built by Anas Ali, a solo founder based in Mandi Bahauddin, Pakistan. It is an early-stage project — actively developed, transparent about limitations, and not VC-funded.

Question 15

Do you have an API?

Accepted Answer

Yes. API access is available on Team and Enterprise plans. Documentation is available at /docs/api.

Question 16

What file types do you support?

Accepted Answer

Text (paste or URL), images (JPG, PNG, WEBP), audio (MP3, WAV, M4A), and video (MP4, MOV, WEBM).

Frequently Asked Questions

Pricing & Plans

Verification & Accuracy

Privacy & Security

API & Integration

Still have questions?