AudioWeaver - Audio Extractor
GRATUIT
Free
Utilities
Free
Slop Or Not - AI Detector
by Numen Technologies Limited
Description
Spot AI-generated writing and images privately on your iPhone — entirely on the Apple Neural Engine. Nothing gets uploaded. Nothing leaves your device.
Most AI scanners : Winston AI, Originality, Copyleaks, Turnitin, Hive, the rest send your content to remote servers. Some also share it with third parties like advertisers and cloud providers. Slop Or Not doesn't. If you're checking for plagiarism or reviewing something sensitive and can't risk it sitting on someone else's server, that's the point.
How accurate is it?
C2PA / IPTC watermarked images: 100% detection.
AI-generated text: 95% on mixed-domain benchmarks (comparable to GPTZero).‡
AI-generated deepfake images: 94% on Mirage-Test - up 18 points from our previous model.
What else does it do?
A text cleanup tool strips invisible unicode, suspicious punctuation, and homoglyphs that AI text sometimes carries. It also catches paraphrase patterns from humanizer tools. One tap.
Readability scoring rates how easy your text is to read. Flesch Reading Ease and Flesch-Kincaid for English, GULPEASE for Italian, LIX for Swedish, Danish, and Norwegian, Flesch-Szigriszt for Spanish, and Wiener Sachtextformel for German.
Works offline after the initial model download.
Comes with a sticker pack for calling out AI slop in group chats.
How does it work?
Everything runs locally on your device. Models download once during setup. Paste text or pick photos to verify. Real results in seconds - all processed on-device via the Apple Neural Engine.
Pricing
Free: 3 checks/day. Hit a 5-day streak and it goes to 5/day.
Pro: unlimited. Monthly, annual, or lifetime. Annual and lifetime support Family Sharing.
Background
We shipped Private LLM almost three years ago - first on-device LLM app for iPhone. On-device AI is what we do.
*Image accuracy: Mirage-Test benchmark, March 2026. Text accuracy: internal mixed-domain benchmarks. Results may vary with newer models and adversarial attacks.
‡ English only for now. More languages coming.
Internet needed only for initial model downloads and occasional model updates and model decryption key downloads from the App Store. All detection runs on-device.
Sticker packs are for educational and informational purposes. Please follow your local data-handling regulations.
Terms (EULA): https://www.apple.com/legal/internet-services/itunes/dev/stdeula/
Privacy Policy: https://slopornot.ai/privacy-policy
Most AI scanners : Winston AI, Originality, Copyleaks, Turnitin, Hive, the rest send your content to remote servers. Some also share it with third parties like advertisers and cloud providers. Slop Or Not doesn't. If you're checking for plagiarism or reviewing something sensitive and can't risk it sitting on someone else's server, that's the point.
How accurate is it?
C2PA / IPTC watermarked images: 100% detection.
AI-generated text: 95% on mixed-domain benchmarks (comparable to GPTZero).‡
AI-generated deepfake images: 94% on Mirage-Test - up 18 points from our previous model.
What else does it do?
A text cleanup tool strips invisible unicode, suspicious punctuation, and homoglyphs that AI text sometimes carries. It also catches paraphrase patterns from humanizer tools. One tap.
Readability scoring rates how easy your text is to read. Flesch Reading Ease and Flesch-Kincaid for English, GULPEASE for Italian, LIX for Swedish, Danish, and Norwegian, Flesch-Szigriszt for Spanish, and Wiener Sachtextformel for German.
Works offline after the initial model download.
Comes with a sticker pack for calling out AI slop in group chats.
How does it work?
Everything runs locally on your device. Models download once during setup. Paste text or pick photos to verify. Real results in seconds - all processed on-device via the Apple Neural Engine.
Pricing
Free: 3 checks/day. Hit a 5-day streak and it goes to 5/day.
Pro: unlimited. Monthly, annual, or lifetime. Annual and lifetime support Family Sharing.
Background
We shipped Private LLM almost three years ago - first on-device LLM app for iPhone. On-device AI is what we do.
*Image accuracy: Mirage-Test benchmark, March 2026. Text accuracy: internal mixed-domain benchmarks. Results may vary with newer models and adversarial attacks.
‡ English only for now. More languages coming.
Internet needed only for initial model downloads and occasional model updates and model decryption key downloads from the App Store. All detection runs on-device.
Sticker packs are for educational and informational purposes. Please follow your local data-handling regulations.
Terms (EULA): https://www.apple.com/legal/internet-services/itunes/dev/stdeula/
Privacy Policy: https://slopornot.ai/privacy-policy
What's new (v1.0.9)
- Text cleanup now offers original, cleaned, and changes views
- New settings toggle to enable/disable Neural Engine inference for the ai generated image detection model.
- Minor bug fixes and updates
Thank you for choosing Slop Or Not. We are committed to continue improving the app and to making it more useful for you. For support requests and feature suggestions, please feel free to join our Discord, email us at support@slopornot.ai, or tweet us @SlopOrNotAI. If you enjoy the app, leaving an App Store review is a great way to support us.
- New settings toggle to enable/disable Neural Engine inference for the ai generated image detection model.
- Minor bug fixes and updates
Thank you for choosing Slop Or Not. We are committed to continue improving the app and to making it more useful for you. For support requests and feature suggestions, please feel free to join our Discord, email us at support@slopornot.ai, or tweet us @SlopOrNotAI. If you enjoy the app, leaving an App Store review is a great way to support us.
