MIT AI Risk Repository · Risk Sub-Category · 18.04.01
Influence operations
Category: Malicious Use
Description
"Facilitating large-scale disinformation campaigns and targeted manipulation of public opinion"
From Sociotechnical Safety Evaluation of Generative AI Systems (Weidinger2023), as extracted by the MIT AI Risk Repository (CC BY 4.0).
Classification
- Domain
- 4. Malicious actors
- Causal entity
- Human
- Intent
- Intentional
- Timing
- Post-deployment
Subdomain definition: Using AI systems to conduct large-scale disinformation campaigns, malicious surveillance, or targeted and sophisticated automated censorship and propaganda, with the aim to manipulate political processes, public opinion and behavior.
Real-world incidents in this subdomain
- Suspected AI-Generated Deepfake Video Reportedly Targeted Former Chhattisgarh Chief Minister Bhupesh Baghel on Instagram
- Purportedly AI-Manipulated Satellite Image Reportedly Claimed Iranian Strike Destroyed U.S. Radar in Qatar
- Purported AI-Generated War Footage Reportedly Circulated Widely Online During the Opening Phase of the War in Iran
- Network of Allegedly Fake Facebook Profiles with Purportedly AI-Generated Images Amplified Posts by Bulgaria's 'There Is Such a People' (ITN) Party
- Purportedly AI-Generated Image Reportedly Circulated Ahead of Thai Election Depicting PM Anutin Charnvirakul Dining with Benjamin Mauerberger
- Purportedly AI-Altered Images Reportedly Distort Evidence After Minneapolis Shooting of ICU Nurse Alex Pretti
How other frameworks describe this risk
Other entries from Weidinger2023
- Representation & Toxicity Harms
- Unfair representation
- Unfair capability distribution
- Toxic content
- Misinformation Harms
- Propagating misconceptions/ false beliefs
- Erosion of trust in public information
- Pollution of information ecosystem
- Information & Safety Harms
- Privacy infringement
- Dissemination of dangerous information
- Malicious Use