Fact · Voice privacy and security
hundredsverbatim text sequences
The Carlini et al. study recovered hundreds of verbatim text sequences from the training data of GPT-2.
Checked against its source on October 17, 2026
In the source's own words
“are able to extract hundreds of verbatim text sequences from the model's training data”
Extracting Training Data from Large Language Models, Nicholas Carlini et al., 30th USENIX Security Symposium, 2021, Abstract.
Open the original- What it measures
- The extracted sequences included public personally identifiable information (names, phone numbers and email addresses), IRC conversations, code and 128-bit UUIDs.
- What it does not show
- This is a result on a model built on web text. It does not measure how much private speech a dictation model keeps.