Applications

Can GPT-3 write misinformation? Yup, it sure can

Deepnews.ai
| June 3, 2021

Andrew Lohn discusses the different ways CSET used GPT-3 to automate misinformation.

How easy is it for AI-written disinformation to fool readers?

KCBS Radio
| May 28, 2021

CSET Senior Fellow Andrew Lohn shares his research using GPT-3 to write disinformation campaigns.

AI could scale up disinformation campaigns, researchers warn

Global Government Forum
| May 27, 2021

CEST's use of GPT-3 reveals an inevitable future where automated messages can create disinformation campaigns online.

Do Tech Companies’ AI Investments Actually Overlap with National Priorities?

Federal News Network
| May 26, 2021

CSET's Tim Hwang was invited to join the Federal Drive podcast to discuss tech companies' investments in AI and its misalignment with national priorities.

Shared tactics between emergency response and IT defense

GCN
| May 24, 2021

CSET Senior Fellow Andrew Lohn weighs the strengths and weakness of AI used in cybersecurity.

Can AI write believable misinformation?

Government Technology
| May 24, 2021

Using GPT-3, CSET was able to generate written messages of misinformation disguised as a human.

AI Can Write Disinformation Now—and Dupe Human Readers

Wired
| May 24, 2021

Using an AI algorithm called GPT-3, CSET was able to generate misinformation.

Report Highlights How AI Could Amplify Future Disinformation Campaigns

Nextgov
| May 21, 2021

CSET report examines how GPT-3, a new AI system, has the capability to automate the future of disinformation campaigns.

The Next QAnon Conspiracy Could Be Created By Robots

Vice
| May 20, 2021

Vice featured CSET's report, "Truth, Lies, and Automation," which discusses how language processing models could be used to automate and fuel disinformation campaigns.

Behind India’s AI Patent Boom

Analytics India Magazine
| May 20, 2021

Analytics India highlighted CSET research studying the AI patent boom in India and around the world.