Forecasting potential misuses of language models for disinformation campaigns and how to reduce risk
OpenAI just published the playbook on how bad actors will weaponize LLMs against your information ecosystem—and how to stop them.
Why it matters
OpenAI researchers released a comprehensive framework for understanding and mitigating LLM-enabled disinformation risks. This is a public governance/safety research contribution that matters to CTOs, security leaders, and policy teams navigating AI risk.
The key facts
10 to knowCollaboration: OpenAI, Georgetown CSET, Stanford Internet Observatory
30 participants in October 2021 workshop: disinformation researchers, ML experts, policy analysts
Over 1 year of research timeline
Report published January 2023
Framework focuses on threat modeling and mitigation strategies for LLM-augmented disinformation
OpenAI collaborated with Georgetown CSET and Stanford Internet Observatory
October 2021 workshop with 30 disinformation researchers, ML experts, and policy analysts
Report based on >1 year of research
Framework provided for analyzing mitigations to disinformation risks
Published January 11, 2023
Go to the source
OpenAI Blogopenai.com
Publisher excerpt: OpenAI researchers collaborated with Georgetown University’s Center for Security and Emerging Technology and the Stanford Internet Observatory to investigate how large language models might be misused for disinformation purposes. The collaboration included an October 2021 workshop bringing together…