Marketing & Comms

Brand safety AI rollout at Coca‑Cola sparks internal pushback

Nolan Howard · 6 Sep 2026

The new AI‑driven brand safety platform that Coca‑Cola started deploying across its social channels last month is flagging 30 % more user‑generated posts than our previous manual process. We've seen the compliance team spending extra hours on each flagged post, which means campaign approvals are slipping and media budgets are getting held up. Agencies are pushing back, saying the tool over‑blocks content that aligns with our creative briefs, and they're threatening to pause future collaborations. How are other marketers balancing strict AI enforcement with the need for agile, on‑time campaign launches?

👍 680 💬 3

3 replies

Jesse Martinez · 6 Sep 2026

My gut says the AI is just a shiny distraction. If you set a confidence threshold you could let the compliance team only review the top‑risk 10 % of flags. We've tried a similar setup with Sprinklr and saw a 20 % drop in manual reviews without hurting brand safety.

Adrian Ford · 6 Sep 2026

Honestly I rushed the AI rollout and skipped the final agency review, which meant the thresholds were set too tight. Now the compliance team is drowning in false positives and the media spend is stuck, something I could have avoided by running a pilot with Sprinklr’s test mode. Going forward I’ll push for a staggered launch and get the creative leads to sign off before we flip the switch.

Antonio Stone · 6 Sep 2026

Tried adding a quick manual review queue after the AI flagging using Sprinklr, but it just doubled the turnaround time and didn't cut the false‑positive rate.