Important points:
- AI success depends on data, culture, and disciplined experimentation.
- Human creativity remains an irreplaceable advantage of beauty even amidst automation.
- Agenttic commerce redefines personalization, speed, and brand connection.
In today’s rapidly growing virtual world, the question is no longer whether to use AI, but how to use AI. This was a central theme at BeautyMatter’s NEXT50 Summit in Los Angeles, specifically the panel “AI Is Everywhere: Turning Buzz into Business in Beauty,” moderated by John Cafarelli, co-founder and president of BeautyMatter.
Cafarelli grouped Jeffery Wagstaff, CEO of CōsNova USA; Frank Freeman, Vice President, AI and Digital Disruption, Dermalogica and Unilever Prestige. and Chaz Giles, managing partner of Alida Labs.
The conversation quickly moved beyond the hype and headlines, delving deep into the midst of the disruption of AI adoption and where the real business impact of AI is beginning to take shape.
From buzzword to business tool
AI is nothing new to Wagstaff, which oversees a mass-market beauty brand with hundreds of SKUs. It’s a must have. “We see AI as part of a partnership and a member of the team that enables us to be more productive and agile,” he said.
AI has already proven its value in forecasting and inventory management, an area where the right algorithms can save millions of dollars. But Wagstaff makes it clear that the real opportunity lies not only in automation. Reassigning in progress. “AI can help you perform mundane tasks like analyzing large amounts of data, so your team can focus on thinking, strategy, and creativity.”
Issues, culture and infrastructure
At Dermalogica and Unilever Prestige, Freeman takes a systematic approach to innovation. His framework for AI adoption revolves around three deceptively simple questions. “Are we working on the right problem?” Do we have the right culture? Do we have the right infrastructure?
“Cost savings and efficiencies are great,” Freeman said. “But what I’m even more excited about is the increase in sales.”
Culture is just as important as code, he added. “It’s both top-down leadership and bottom-up innovation.” That duality is embodied in Unilever Prestige’s decision to “disable human translation.” This was a bold decision that forced the team to develop an OpenAI-based system that was scalable across brands and markets. “We said, let’s stop human translation. Think about it,” he recalls. “And the team did.”
Another of Freeman’s innovation tenets, “Explore, Assign, Pioneer, Scale,” enables teams to identify problems and experiment with solutions. “We ask our interns what they want them to do, and it starts from there.”
reign before greatness
A commonly shared concern across the panel was governance. Data confidentiality, privacy, and security are key to any discussion of AI in beauty. “We laid the foundation with a governance framework,” Wagstaff said. “We are using AI ethically, effectively and prudently from a consumer perspective.”
Mr. Freeman discussed balancing innovation and compliance, particularly under the European Union’s General Data Protection Regulation (GDPR). “Many organizations are becoming risk-averse, and for good reason. But many companies feel like their hands are tied, and they’re in what I would call a situation.” [Microsoft] Co-Pilot Purgatory, where only the use of co-pilots is allowed, but co-pilots do not function. ”
To avoid paralysis, Unilever Prestige created clean GenAI guardrails. “We’re very specific about what we can’t do. But if you have a use case, let us know and we’ll find a way to make it happen,” Freeman explained.
The Myth of AI Magic
While Wagstaff and Freeman talked from inside enterprise systems, Chaz Giles zoomed out to examine the deeper infrastructure behind successful AI integrations. His message was clear. Without the fundamentals, there is no AI magic. “Almost every problem we’ve ever seen with AI is either data or human, and almost never the AI itself.”
Giles described AI as a tool that exposes vulnerable systems rather than fixing them. “If your data is messy, AI will clean up the mess faster. If your workflow is broken, AI will break it faster.” That’s why his first question for brands exploring AI isn’t about the technology. It’s about people and how they understand technology.
Giles also cautioned against brands trying to build their own models. “No one in this room should be building their own AI models,” he said flatly. “It’s better to buy off-the-shelf technology and customize it to your use case.”
He likened the introduction of AI to electricity in the early 20th century. “Initially, companies hired power managers, but eventually power just became part of business operations. AI is the same way. It’s a strategy, not a separate strategy.”
