SN32 reported 99.8% AI-generated text detection accuracy across four languages, with false-positive rates below 1% on authentic student writing datasets. […]
<p class="wp-block-paragraph"><a href="https://taodaily.io/bittensor-subnet-showcase-sn32-its-ai/">It’s AI (SN32)</a> has reported AI-generated text detection accuracy as high as 99.8% across four languages, alongside false-positive rates below 1% on authentic student writing.</p> <p class="wp-block-paragraph">The results cover Arabic, German, French, and Spanish, demonstrating the subnet’s progress in distinguishing AI-generated content from human writing.</p> <p class="wp-block-paragraph">The findings also highlight its focus on reducing incorrect classifications, particularly when evaluating work produced by real people.</p> <figure class="wp-block-image size-large is-resized"><img fetchpriority="high" decoding="async" width="1024" height="328" src="https://taodaily.io/wp-content/uploads/2026/10/image-257-1024x328.png" alt="" class="wp-image-25898" style="aspect-ratio:3.12;width:624px;height:auto" srcset="https://taodaily.io/wp-content/uploads/2026/10/image-257-1024x328.png 1024w, https://taodaily.io/wp-content/uploads/2026/10/image-257-300x96.png 300w, https://taodaily.io/wp-content/uploads/2026/10/image-257-768x246.png 768w, https://taodaily.io/wp-content/uploads/2026/10/image-257.png 1286w" sizes="(max-width: 1024px) 100vw, 1024px" /><figcaption class="wp-element-caption"><a href="https://its-ai.org/en">It’s AI v. Other Detection Solutions</a></figcaption></figure> <p class="wp-block-paragraph"><a href="https://taodaily.io/why-its-ai-sn32-fits-the-reuters-institutes-vision-for-the-future-of-media/">It’s AI (SN32)</a> allows for contributors to collaborate to develop tools that identify whether text was written by humans or artificial intelligence.</p> <p class="wp-block-paragraph">Its latest evaluation used DetectRL-X (ACL 2026), a multilingual benchmark covering six writing domains, modern AI models, and AI-edited content.</p> <p class="wp-block-paragraph">Approximately 12,000 human-written and AI-generated texts were evaluated per language to measure the detector’s performance across different writing conditions.</p> <figure class="wp-block-image size-large is-resized"><img decoding="async" width="1024" height="576" src="https://taodaily.io/wp-content/uploads/2026/10/image-261-1024x576.png" alt="" class="wp-image-25902" style="aspect-ratio:1.7777777777777777;width:624px;height:auto" srcset="https://taodaily.io/wp-content/uploads/2026/10/image-261-1024x576.png 1024w, https://taodaily.io/wp-content/uploads/2026/10/image-261-300x169.png 300w, https://taodaily.io/wp-content/uploads/2026/10/image-261-768x432.png 768w, https://taodaily.io/wp-content/uploads/2026/10/image-261-678x381.png 678w, https://taodaily.io/wp-content/uploads/2026/10/image-261.png 1672w" sizes="(max-width: 1024px) 100vw, 1024px" /><figcaption class="wp-element-caption">What DetectRL-X (ACL 2026) Is About</figcaption></figure> <p class="wp-block-paragraph">The detector achieved 99.8% accuracy in Arabic, 99.5% in German, 99.4% in French, and 96.0% in Spanish.</p> <p class="wp-block-paragraph">These results demonstrate strong performance across languages with different linguistic structures. By testing the system on multiple types of writing and AI-generated content, the evaluation provides a picture of how effectively SN32 can distinguish between human and machine-generated text in multilingual settings.</p> <figure class="wp-block-image size-large is-resized"><img decoding="async" width="1024" height="347" src="https://taodaily.io/wp-content/uploads/2026/10/image-259-1024x347.png" alt="" class="wp-image-25900" style="aspect-ratio:2.957345971563981;width:624px;height:auto" srcset="https://taodaily.io/wp-content/uploads/2026/10/image-259-1024x347.png 1024w, https://taodaily.io/wp-content/uploads/2026/10/image-259-300x102.png 300w, https://taodaily.io/wp-content/uploads/2026/10/image-259-768x260.png 768w, https://taodaily.io/wp-content/uploads/2026/10/image-259.png 1881w" sizes="(max-width: 1024px) 100vw, 1024px" /><figcaption class="wp-element-caption"><a href="https://its-ai.org/en/accuracy">It’s AI (SN32)’s Arabic Prowess on DetectRL-X</a></figcaption></figure> <p class="wp-block-paragraph">Beyond accuracy, the team examined false positives, which occur when genuine human writing is incorrectly classified as AI-generated.</p> <p class="wp-block-paragraph">Using authentic student writing datasets, It’s AI (SN32) recorded false-positive rates of 0.89% in Arabic, 0.48% in German, 0.56% in French, and 0.72% in Spanish.</p> <p class="wp-block-paragraph">The Spanish evaluation included more than 20,000 texts, providing a substantial sample for assessing incorrect classifications.</p> <figure class="wp-block-image size-large is-resized"><img loading="lazy" decoding="async" width="1024" height="576" src="https://taodaily.io/wp-content/uploads/2026/10/image-260-1024x576.png" alt="" class="wp-image-25901" style="aspect-ratio:1.7777777777777777;width:624px;height:auto" srcset="https://taodaily.io/wp-content/uploads/2026/10/image-260-1024x576.png 1024w, https://taodaily.io/wp-content/uploads/2026/10/image-260-300x169.png 300w, https://taodaily.io/wp-content/uploads/2026/10/image-260-768x432.png 768w, https://taodaily.io/wp-content/uploads/2026/10/image-260-678x381.png 678w, https://taodaily.io/wp-content/uploads/2026/10/image-260.png 1920w" sizes="(max-width: 1024px) 100vw, 1024px" /><figcaption class="wp-element-caption"><a href="https://x.com/ai_detection/status/2114797629658026227">SN32’s Multilingual Accuracy Report on 4 Languages</a></figcaption></figure> <p class="wp-block-paragraph">These results are particularly relevant to education, where incorrectly flagging a student’s work as AI-generated can lead to unfair accusations.</p> <p class="wp-block-paragraph">It’s AI (SN32) also reported strong English-language performance, achieving 98.3% accuracy on the RAID benchmark and ranking first on MGTD, ahead of GPTZero and Originality AI.</p> <figure class="wp-block-image size-large is-resized"><img loading="lazy" decoding="async" width="1024" height="352" src="https://taodaily.io/wp-content/uploads/2026/10/image-258-1024x352.png" alt="" class="wp-image-25899" style="aspect-ratio:2.902325581395349;width:624px;height:auto" srcset="https://taodaily.io/wp-content/uploads/2026/10/image-258-1024x352.png 1024w, https://taodaily.io/wp-content/uploads/2026/10/image-258-300x103.png 300w, https://taodaily.io/wp-content/uploads/2026/10/image-258-768x264.png 768w, https://taodaily.io/wp-content/uploads/2026/10/image-258.png 1847w" sizes="(max-width: 1024px) 100vw, 1024px" /><figcaption class="wp-element-caption">SN32’s English Benchmark Performance </figcaption></figure> <p class="wp-block-paragraph">Together, the findings show progress in building multilingual detection tools that combine high accuracy with low false-positive rates.</p>