This past March, Bill Gates stated that those who excel in the area of AI personal assistants – AI tools that can perform tasks for users such as web searches, managing productivity tools, or shopping online – would lead the next generation. Right now, it’s looking like the retail industry might be the first to realize Gate’s prediction.
Less than a year after the launch of ChatGPT, Cherubic Ventures’ portfolio company, retail SaaS startup 91APP, introduced Asia’s first native retail AI model, Jooii. Think of Jooii as a personal shopping assistant – users just need to input what type of product they want, who it is for, and any specific requirements, and Jooii will provide direct product recommendations.
This is just one example of how AI is transforming the retail industry and becoming an indispensable tool, from front-end customer interactions to back-end workflows.
Walmart is using conversational AI bots for a different type of interaction – negotiating purchase contracts with human suppliers. The goal of using the AI is to provide the retail giant with more contract flexibility and marketing exposure to incentivize suppliers to offer more payment discounts or extended payment terms.
Through a partnership with contract negotiation AI startup Pactum, Walmart is implementing this negotiation bot with suppliers in the United States, Canada, Chile, and South Africa. They have successfully reached deals with nearly 70% of their suppliers, saving an average of 3% in costs. Walmart has indicated it plans to expand this system from supplier negotiations to negotiations on logistics costs.
Retailers are also adopting generative AI to reduce marketing spend.
Fashion e-commerce platform, WeShop, launched an AI commercial photography tool in the first half of this year. Retailers simply need to upload basic product photos and select criteria such as presentation styles, scenes, and AI model nationalities, and the tool automatically generates high-quality product images.
This can save retailers around 2% of GMV in photography expenses, making the AI a powerful tool for businesses looking to expand into overseas markets with limited budgets.
AI is also revolutionizing customer identity identification. Amazon’s recently launched palm scan payment system, Amazon One, can identify a customer’s identity by scanning their palm prints, veins, and other biometric information that has been linked to their Amazon account and credit card. This allows for easy payments, member verification, and age verification, essentially replacing facial recognition.
However, Amazon faced a significant challenge when developing this system: how to improve scanning accuracy with limited palm data. It turned to generative AI to generate millions of palm images with different lighting, poses, and palm prints to train their AI model. The retailer claims that their identity recognition accuracy is 100 times more accurate than scanning two irises.
However, we should consider taking a pause before fully embracing AI assistants and think about the larger quantity of personal data we will need to provide.
We might not yet have AI assistants with the omniscient capabilities of Iron Man’s Jarvis in the real world. But as long as we continue to integrate more consumer and market research data, these kinds of assistants are not far from reality. The question is whether we are ready for that day.