National Law Review
7/20/2026

Every AI Answer Needs a Byline
Short summary
The article argues that quality journalism is critical infrastructure for AI systems, as AI answers depend on fresh, accurate reporting. A study found major AI models give materially wrong answers to US election questions 90% of the time, with ChatGPT citing state-owned media in over half its foreign-policy answers. The author warns that as AI-generated content proliferates, the value of original reporting increases, and underinvesting in journalism will degrade AI trustworthiness.
- •AI models give wrong answers to US election questions 90% of the time per a cited study
- •ChatGPT cited state-owned media in over half its foreign-policy answers
- •Quality journalism is essential infrastructure for trustworthy AI outputs
Generated with AI, which can make mistakes.
Is this a good recommendation for you?



