A recent report by UNESCO cautioned that not only can Generative AI enable malicious actors to seed disinformation and hate-fueled narratives, but it can also inadvertently invent false or misleading content about historical records.
- Generative AI is an artificial intelligence technology that automatically generates content in response to prompts written in natural-language conversational interfaces.
Concerns highlighted in the report
- AI automated content may invent facts: Generative AI risks spreading distorted and offensive representations of facts which can impede public understanding of historical events.
- AI models can be manipulated to spread hate speech: AI systems may inherit human biases, potentially misrepresenting information about specific events or societal groups, reinforcing prejudices.
- Violation of privacy: AI limits individuals’ ability to control what information about their past is retrieved.
- Lack of transparency and algorithmic biases: It is often unclear which data AI systems have been trained on or how algorithms decide what content to retrieve, generate, or prioritize.
Key Recommendations
- Monitoring and evaluating standards for AI systems’ management of information: Emphasize on preventing prioritized retrieval or generation of content that may be damaging to rights and reputations of individuals.
- Integrating ethical values into AI design: Integrate ethical impact assessment – fairness, non-discrimination and diversity – in AI system design.
