ML at @GPTZeroAI Alumni @rlai_lab @AmiiThinks supervised by @CsabaSzepesvari
Toronto, Ontario
- Tweets 8
- Following 106
- Followers 34
- Likes 51
Pinned Tweet
LLMs are evolving fast. GPTZero is faster.
After tons of research and evaluation, I’m happy to share that GPTZero 4o, our most accurate AI detector, is live now!
GPTZero 4o achieved over 99% average recall across all major LLMs, while keeping false positives under 0.03%
2
2
12
2,013
Love the authenticity and transparency GPTZero preserves. Great work, team!
Here’s our full GPTZero investigation: gptzero.me/news/words-from-t…
We’re also releasing our full dataset publicly here, welcoming further analysis: docs.google.com/spreadsheets…
1
45
LLMs are evolving fast. GPTZero is faster.
After tons of research and evaluation, I’m happy to share that GPTZero 4o, our most accurate AI detector, is live now!
GPTZero 4o achieved over 99% average recall across all major LLMs, while keeping false positives under 0.03%
2
2
12
2,013
New LLMs keep coming out, and @GPTZeroAI is ready for them!
Stay tuned for our results on the most recent set of LLMs that came out in the past week.
2
1
71
And check out our full GPTZero 4o highlights here
After months of heads-down grinding, I’m proud to share that the @GPTZeroAI team has released the world’s most accurate AI detection model, GPTZero 4o.
It’s called 4o because there are four zeros: we mark less than one in 10000 human texts as AI, evaluated across diverse passages of student writing. At the same time, we detect Claude Fable with 98.4% accuracy.
We outperform Pangram in 3rd party benchmarks. Along the way, our research found major biases in the Pangram detector.
1
103
