Machine Learning Times
Machine Learning Times
EXCLUSIVE HIGHLIGHTS
MLW Preview Video: Sarah Kalicin, Data Scientist at Intel Corporation
 In anticipation of her upcoming keynote presentation at Predictive...
MLW Preview Video: Praneet Dutta, Senior Research Engineer at DeepMind
 In anticipation of his upcoming presentation at Deep Learning...
MLW Preview Video: Dean Abbott, President at Abbott Analytics
 In anticipation of his upcoming pre-conference workshop and presentations...
MLW Preview Video: Evan Wimpey, Director of Strategic Analytics at Elder Research
 In anticipation of his upcoming presentation at Predictive Analytics...
SHARE THIS:

4 months ago
OpenAI’s Attempts to Watermark AI Text Hit Limits

 
Originally published in TechCrunch, Dec 22, 2022.

It’s proving tough to rein in systems like ChatGPT.

Did a human write that, or ChatGPT? It can be hard to tell — perhaps too hard, its creator OpenAI thinks, which is why it is working on a way to “watermark” AI-generated content.

In a lecture at the University of Texas at Austin, computer science professor Scott Aaronson, currently a guest researcher at OpenAI, revealed that OpenAI is developing a tool for “statistically watermarking the outputs of a text [AI system].” Whenever a system — say, ChatGPT — generates text, the tool would embed an “unnoticeable secret signal” indicating where the text came from.

OpenAI engineer Hendrik Kirchner built a working prototype, Aaronson says, and the hope is to build it into future OpenAI-developed systems.

“We want it to be much harder to take [an AI system’s] output and pass it off as if it came from a human,” Aaronson said in his remarks. “This could be helpful for preventing academic plagiarism, obviously, but also, for example, mass generation of propaganda — you know, spamming every blog with seemingly on-topic comments supporting Russia’s invasion of Ukraine without even a building full of trolls in Moscow. Or impersonating someone’s writing style in order to incriminate them.”

To continue reading this article, click here.

7 thoughts on “OpenAI’s Attempts to Watermark AI Text Hit Limits

Leave a Reply