Many observers of the current explosion of generative artificial intelligence worry about its impact on our information environment, with concerns being raised about the increasing quantity, quality and personalization of misinformation.
- Increasing amount of misinformation
Explanation of the claim: Due to their ease of access and use, generative AIs can be used to create misinformation/disinformation at scale at low or no cost to individuals and organized actors.
Supposed effect: The increase in the amount of misinformation allows malicious actors to “flood the space” with inaccurate or misleading information, thereby drowning out factual content and/or sowing confusion.
2. Increasing the quality of misinformation
Explanation of the claim: Due to their technical capabilities and ease of use, generative AIs can be used to create higher-quality disinformation.
Supposed effect: Increasing the quality of misinformation leads to increased persuasive potential, as it creates content that is more credible and harder to verify. This would either allow the spread of false information or contribute (with the increased quantity of misinformation) to an epistemic crisis, a general loss of trust in all types of news.
3. Increasing personalization of misinformation
Explanation of the claim: Due to their technical capabilities and ease of use, generative AIs can be used to create high-quality misinformation personalized to a user's tastes and preferences.
Supposed effect: Increasing consumer belief in misinformation, with the same results as above.
4. Involuntary generation of convincing but false information
Explanation of the claim: Generative AIs can generate useful content (e.g., chatbots that generate code). However, they can also generate convincing-looking information that is completely inaccurate. Thus, users can generate misinformation, which can potentially spread, without intending to.
Supposed effect: Misinforming users of generative AI and potentially those with whom they share information
Full article at misinforeview.hks.harvard.edu