Making disinformation cheaper and more effective

Overview

A source-backed snapshot of this AI risk.

Techniques2Attack methods connected to this risk.

Mitigations2Defenses that may help with related attacks.

Records2Source records unified into this concept.

Risk profile

How the MIT AI Risk Repository categorizes this risk.

Domain4. Malicious Actors & Misuse

Subdomain4.1 > Disinformation, surveillance, and influence at scale

Entity1 - Human

Intent1 - Intentional

Timing2 - Post-deployment

CategoryMalicious Uses; Risk area 4: Malicious Uses

SubcategoryMaking disinformation cheaper and more effective

Merged risk records

Source records unified into this canonical risk concept.

2 recordsView all →

MITRISK-Weidinger2021-17.04.01 - Making disinformation cheaper and more effective

"LMs can be used to create synthetic media and ‘fake news’, and may reduce the cost of producing disinformation at scale (Buchanan et al., 2021). While some predict that it will be cheaper to hire humans to generate disinformation (Tamkin et al., 2021), it is possible that LM-assisted content generation may offer a cheaper way of generating diffuse disinformation at scale."

Domain4. Malicious Actors & MisuseSubdomain4.1 > Disinformation, surveillance, and influence at scaleSourceEthical and social risks of harm from language modelsYear2021

MITRISK-Weidinger2022-16.04.01 - Making disinformation cheaper and more effective

"While some predict that it will remain cheaper to hire humans to generate disinformation [180], it is equally possible that LM- assisted content generation may offer a lower-cost way of creating disinformation at scale."

Domain4. Malicious Actors & MisuseSubdomain4.1 > Disinformation, surveillance, and influence at scaleSourceTaxonomy of Risks posed by Language ModelsYear2022