AI Slop and the Destruction of Knowledge
10 days ago
- #Misinformation
- #Scientific Integrity
- #AI in Science
- The author encountered AI-generated definitions on ScienceDirect that were misleading and incorrect, contrary to how cognitive scientists use the term 'domain-general' cognition.
- Concerns were raised about the impact of AI-generated content on scientific integrity, knowledge infrastructure, and the potential harm to students, researchers, and the public.
- The author communicated with ScienceDirect, questioning the inclusion of AI-generated summaries in scientific articles without authors' consent and the spread of misinformation.
- ScienceDirect initially defended the use of AI-generated content, claiming it was expert-written, but later acknowledged the mistake and promised to review and remove problematic definitions.
- The author advocates for the removal of AI features from ScienceDirect to protect the integrity of scientific knowledge and prevent the spread of misinformation.
- References and open letters were shared to encourage resistance against the uncritical adoption of AI in academia and education.