Is GPT-5.6 (Luna, Sol, Terra) Content Detectable?

On July 9, 2026, OpenAI released GPT-5.6, which it describes as the “Frontier intelligence that scales with your ambition.”

‍The GPT-5.6 family of models includes:

• Sol: the new flagship model
• Terra: described as a balanced model suitable for everyday tasks
• Luna: the ‘most cost-efficient model’

OpenAI also notes that the new, highly capable GPT-5.6 suite of models was released with the “most robust safeguards to date.” 

Now that the GPT-5.6 models have been released, the question remains — is GPT-5.6 content still detectable?

As with other new LLM model releases, such as GPT-5.2 and GPT-5.5, we ran a study to find out! 

Read our analysis to find out if GPT-5.6 is still detectable by Originality.ai.

The Quick Answer — Is GPT-5.6 Text Detectable?

Yes, Originality.ai is highly accurate at detecting GPT-5.6 text.

‍GPT-5.6 Luna

• Lite = 96% accuracy
• Turbo = 97% accuracy
• Academic = 97% accuracy
• AI Allowance (at 15% default settings) = 99%+ accuracy

‍GPT-5.6 Sol

• Lite = 97% accuracy
• Turbo = 99% accuracy
• Academic = 97% accuracy
• AI Allowance (at 15% default settings) = 99%+ accuracy

‍GPT-5.6 Terra

• Lite = 97% accuracy
• Turbo = 98% accuracy
• Academic = 97% accuracy
• AI Allowance (at 15% default settings) = 99%+ accuracy

The GPT-5.6 Detectability Study

AI companies are continuously innovating and releasing new LLMs. Just consider the recent releases from OpenAI, like GPT-5.5, Anthropic’s Claude Opus 4.7, Gemini 3 from Google, or Grok 4.1 Fast from xAI.

With the rapid development and release of new AI models, it’s critical for AI detectors to test accuracy at identifying the latest LLMs. 

So, to find out whether Originality.ai can still detect AI content generated by GPT-5.6, we ran a test as soon as it was released.

The GPT-5.6 Detectability Test 

Although our AI detection models were not trained on GPT-5.6 data, this study assesses their capability to identify text generated by GPT-5.6.

‍The Dataset:

‍5,000 high-quality human-written samples spanning:

• Domains/Topics: health, news, finance, technology, arts & entertainment
• Content Types: blogs, reviews, expository articles, conversational text
• Prompt Structures: to maximize stylistic and topical diversity

Then, GPT-5.6 was used to generate completions for each human-written sample prompt. 

This produced a balanced dataset to evaluate the accuracy of Originality.ai’s AI detection capabilities.

The GPT-5.6 Detectability Results

After testing the Originality.ai AI detection models on this dataset, we found that GPT-5.6 is still highly detectable by the industry-leading Originality.ai AI detector.

‍These are the accuracy ratings for GPT-5.6 by Originality.ai AI detection models:

‍

Final Thoughts

Although GPT-5.6 didn’t yet exist when the Originality.ai AI detection models were trained, Lite, Turbo, and Academic show a strong generalization performance and achieved an accuracy of 96%+.

‍AI Allowance, the latest model from Originality.ai, achieved 99%+ accuracy at default settings (15% AI allowance).

‍So, yes — GPT-5.6 is still highly detectable by Originality.ai!

For writers, editors, students, and teachers, Originality.ai is still highly effective at identifying AI text from the latest LLMs, so you can maintain transparency and find out the authorship of the content you’re reading.

Try Originality.ai’s AI detector today!

‍Learn more about AI detection accuracy:

• AI Detection Accuracy Study
• 3rd Party AI Detection Accuracy Studies: Meta-Analysis

‍AI studies for recently released models:

• Gemini 3
• Grok 4.1 Fast
• GPT-5.3 Instant and GPT-5.4
• Claude 4.7 Opus
• Llama 4 Maverick and Llama 4 Scout
• DeepSeek