Forecasting potential misuses of language models for disinformation campaigns and how to reduce risk
Researchers from OpenAI, Georgetown University's Center for Security and Emerging Technology, and Stanford Internet Observatory collaborated on a study to identify how large language models could be misused in disinformation campaigns. They brought together experts from various fields, including machine learning and policy analysis, in an October 2021 workshop. The resulting report outlines potential threats and proposes a framework for mitigating these risks.
Researchers from OpenAI, Georgetown University's Center for Security and Emerging Technology, and Stanford Internet Observatory collaborated on a study to identify how large language models could be misused in disinformation campaigns. They brought together experts from various fields, including machine learning and policy analysis, in an October 2021 workshop. The resulting report outlines potential threats and proposes a framework for mitigating these risks.
---
Why it matters: This research matters because it highlights the potential for large language models to be exploited in malicious ways, which could have significant consequences for public discourse and trust in information sources.
Source: https://openai.com/index/forecasting-misuse
This article was originally published at: https://openai.com/index/forecasting-misuse