
AI models require independent, expert-led audits to ensure accuracy in high-stakes domains like politics, healthcare, and personal finance. Currently, AI companies often grade their own performance, creating a trust gap that necessitates external accountability similar to industry-standard safety testing. While social media platforms previously prioritized engagement, enterprise-focused AI models are increasingly incentivized to optimize for factual precision. Campbell Brown, founder of Forum AI and former head of news at Meta, emphasizes that relying on domain-specific experts—rather than engineers—to architect benchmarks is essential for scaling intelligence. As legacy news organizations face declining trust and existential threats, they must adapt to a landscape where younger audiences increasingly rely on chatbots and social media for information. Proactive, entrepreneurial approaches to these technological shifts remain the only viable path for sustaining credible journalism in an era of rapid AI integration.
Sign in to continue reading, translating and more.
Open full episode in Podwise