---
id: "open-question-modality-vs-content"
type: "open-question"
source_timestamps: ["§ The Road Ahead"]
tags: ["academic-research", "methodology-evaluation"]
related: ["claim-verbal-vs-typed-responses", "action-establish-metrics", "concept-scaled-empathy"]
resolutionPath: "Controlled academic studies isolating the variables of modality (voice/text, human/AI avatar) and content (static script vs. dynamic LLM probing) to measure their independent effects on response quality."
sources: ["commercial"]
sourceVaultSlug: "hbr-seg-commercial"
originDay: 5
articleStem: "hbr-new-30-ai-scale-customer-research"
sourceUrl: "https://hbr.org/2026/04/how-ai-helps-scale-qualitative-customer-research"
sourceTitle: "How AI Helps Scale Qualitative Customer Research"
---
# Source of AI Moderation Gains: Modality vs. Content

**Open question.** Academic research is ongoing to determine exactly *why* AI moderation yields better or different results. It is unclear how much the gains come from the **modality** of interaction (AI vs. human; voice vs. text) versus the **content** of the interaction (the LLM's ability to dynamically adjust questions and probe deeper).

This directly qualifies [[claim-verbal-vs-typed-responses]] (is the 7× a voice effect, a dynamic-probing effect, or both?) and [[concept-scaled-empathy]].

**Resolution path.** Controlled studies that **isolate variables** — modality (voice/text, human/AI avatar) and content (static script vs. dynamic LLM probing) — to measure their independent effects on response quality. This is exactly the kind of rigor called for in [[action-establish-metrics]].