Session

The Current State of AI Generated Text Detection

New articles are being released frequently discussing enhancements in language models. Models like GPT2 and BERT are able to mimic human text where even humans have a hard time distinguishing that the text was generated by a machine. As these models continue to improve, the ability for them to be used in a malicious way is increasing. Bad actors could use these models to automate phishing emails, impersonate a celebrity on twitter, or spread fake news. It is important that security researchers look to develop methods to detect machine generated text as a defense against these use cases. GROVER and GLTR are two state of the art frameworks in machine generated text detection. McAfee is working to utilize these frameworks as a way to detect machine generated text.

Biography

Latest news

Podcast: Avast’s Michal Pechoucek on what e...

Michal is one of the chief architects behind CyberSec&AI Connected, which takes place online on ...

Bobby Filar, Lead Data Scientist at Elastic, ...

Bobby Filar is the Lead Data Scientist at Elastic where he employs machine learning and natural lang...

Professor Lorenzo Cavallaro on adversarial ma...

Lorenzo leads the Systems Security Research Lab where he specializes in the intersection of program ...

McAfee’s Celeste Fralick talks algorithm bi...

Celeste Fralick is the Senior Principal Engineer and Chief Data Scientist for McAfee. She is respons...