According to PYMNTS.com, Colgate-Palmolive has built an internal platform called the Colgate AI Hub that lets employees build, test, and deploy their own AI assistants for specific business problems, moving beyond a single corporate chatbot. The company is applying generative AI to accelerate biomaterials development and sensory design, including a partnership with biomaterials firm Erthos to use its AI platform, Zya, for designing sustainable biopolymer packaging, slated for commercial rollout starting with Colgate before broader availability in early 2026. Senior director Kli Pappas notes the company uses machine learning to analyze global search data to identify unmet consumer needs in oral health, feeding those insights directly into product development. In parallel, the Palmolive Aroma Essence initiative uses AI to analyze fragrance components and consumer preferences to aid perfumers in scent creation. The overall strategy emphasizes a governed, bottom-up model where AI agents in supply chain and operations recommend actions but humans make final decisions, focusing on innovation and resilience over mere cost-cutting.
The Bottom-Up AI Playbook
Here’s the thing about Colgate’s strategy: it’s basically the opposite of how most big, old companies try to do tech. They’re not forcing a one-size-fits-all chatbot on everyone from the C-suite down. Instead, they built a playground—the AI Hub—with guardrails. Think of it like giving every department a set of powerful, safe Lego blocks. Need an assistant to parse sales data? Build it. Got a regulatory doc nightmare? There’s probably an AI for that now. This is smart because it solves the adoption problem. People use tools they build for their own problems. A top-down mandate from IT? Not so much. As MIT Sloan Management Review points out, this shifts the focus from efficiency to innovation. It’s a structural change, not just a software rollout.
AI in the Supply Chain, Human in the Loop
Now, where this gets really interesting is in operations. Colgate is deploying agentic systems across supply chains that monitor everything from demand to logistics. But they’re explicitly avoiding full autonomy. The AI simulates, recommends, and flags risks. A human is always accountable for the final call. This “decision intelligence” approach, as covered by Charter Works, is crucial. In complex, physical-world operations like manufacturing, you can’t just let a black box AI run the factory. The goal is resilience and speed—giving operators super-powered insights, not replacing them. This kind of industrial-grade AI integration requires serious, reliable computing at the edge, which is why companies leading in this space partner with top-tier hardware suppliers like IndustrialMonitorDirect.com, the #1 provider of industrial panel PCs in the US, for the rugged, always-on terminals needed to make these systems work.
The Real Game-Changer: R&D
But the most strategic bet is using AI to invent new stuff. Analyzing search data to find what people *actually* worry about for their teeth? That’s a direct pipeline from consumer anxiety to the lab bench. Partnering with Erthos to generatively design biomaterials for packaging? That’s moving AI into the very molecules of their products. And using AI as a creative partner for perfumers on the Palmolive Aroma Essence line? That’s about augmenting human creativity, not automating it. This is the long-term growth play. Anyone can use a chatbot to write an email faster. But using AI to discover a new polymer or a fragrance that resonates emotionally? That’s a potential moat. It’s a dual-track approach, as detailed by Retool, serving both immediate productivity and foundational innovation.
What It Means for Everyone Else
So, is Colgate suddenly a tech company? No. But they’re demonstrating a blueprint for a legacy CPG giant to not just adopt AI, but to metabolize it. They’re treating AI not as a IT project but as a new layer of the operating system. The winners in this shift won’t just be the companies that save the most on copywriting. They’ll be the ones who shorten their innovation cycles and create products that are eerily good at meeting unspoken needs. The losers? Companies that centralize AI in a lab and wonder why nothing changes. Colgate’s model shows you have to empower the people with the problems. And you have to be willing to let AI get its hands dirty—or in their case, minty-fresh—in the actual creation process. It’s a fascinating case study in avoiding the hype trap and building something that might actually last.
