Francesca Roche sits down with Grace Putney, Director of Client Success at ICUC, to examine why social media community management is becoming a bigger customer experience challenge as brands process expanding volumes of customer data.
New ICUC data shows that 46% of brand-to-consumer social interactions are either handled entirely by people or use automation to classify, tag, and route the issue before a person makes the final decision.
For CX leaders under pressure to automate, the finding raises an important question about where efficiency ends and accountability begins.
“AI should be used for efficiency and humans should be used for empathy,” Putney told CX Today.
You need judgment and nuance and you need to be empathetic with people when they have encountered an issue.”
AI can help brands process large volumes of inquiries and customer signals. It can identify emerging themes, categorize requests, and direct issues to the appropriate team.
That capability can create more capacity for community managers, particularly when social channels move quickly and customer expectations for a response remain high.
However, the interview highlights why routing an issue is not the same as resolving it. A customer comment may point to a repeat problem, require a sensitive response, or signal a concern that should reach a broader team.
Putney explains that human review helps brands assess the context and potential consequences behind each interaction.
The risks become more acute as AI-generated spam grows more sophisticated. Reddit says it blocks around 23MN spam views per day, yet users continue to report seeing AI-generated spam.
Community management teams must now assess content that can appear more credible than the obvious bots and scams they previously encountered.
“You’re having to use finer judgment,” Putney says.
“It’s coming at scale.”
The right approach also changes by platform. ICUC’s report found that YouTube requires human moderation on 14% of public interactions, compared with less than 1% on X.
Putney points to YouTube’s longer content and conversation lifecycles, where an older video can resurface and reignite audience engagement long after publication.
For brands, this makes a uniform social moderation model difficult to sustain. Different audiences, formats, and customer behaviors need different levels of oversight.
Putney also addresses the temptation to see AI as a simple cost-cutting tool. “The challenge or the mistake is if you’re using AI as a replacement instead of an aid or an amplifier,” she says.
Watch the full interview to hear how leaders can apply governance, guardrails, and human expertise as AI becomes more embedded in social CX.
