This week, I sat down with Justin Honaman, the Global Head of Retail, Restaurants, and Consumer Goods at AWS, to discuss how the world’s biggest brands are moving past the hype and into production with Agentic AI.
3 Big Insights from the AI Frontlines
Honaman notes that Agentic AI is moving even faster than the pace of GenAI. While many are still processing the basics of generative technology, Honaman highlights that the real shift is toward agents that do not just talk but actually act and collaborate. AWS demonstrated this capability at NRF by showing agents handling market research, product design, bill of materials, and supply chain planning in real time.
Honaman explicitly stated that the days of needing all data in one big platform are over because modern technology can now connect across different platforms and clouds.He notes, however, that it has to be clean, connected, and accessible before any AI can work.
The Rise of "Buy For Me": One of the most fascinating agentic use cases Honaman highlighted is Amazon’s "Buy For Me" function. If a shopper searches Amazon for a product that isn't sold there, an AI agent (powered by Amazon Nova, Claude, and Bedrock) can autonomously navigate to the brand's own website to find and complete the purchase.
The Highlight
The Reality Check: You can't reach Agentic status with "messy" data. But the good news is that you do not have to do a giant migration. Honaman explains how to harness data in a more approachable way.
