- Salesforce demonstrated its new AI platform "Einstein" at a recent event, showcasing the platform's potential usage for inventory management, Supply Chain 247 reported.
- Coca-Cola, a longtime client of Salesforce, allowed one of its coolers, stocked with product, to serve as Einstein's foil. The visual recognition engine easily accounted for the varieties of Coca-Cola stock within the forefront of the cooler display cabinet.
- Salesforce chief scientist and Einstein developer Richard Socher noted that combining this AI with the company's forecasting models, the entire restocking process can become automated and lead to the AI driving the business decision-making.
AI technology offers the tremendous advantage of being able to perform rote, repetitive tasks that are common in inventory management. However, despite the promise of such AI tech as Einstein, there is still a literal learning curve necessary, and it is easy to overlook or forget its limitations, when in fact, they are significant hurdles in the near term.
While it's unlikely that Einstein will be restocking and managing soda coolers immediately, its built-in learning capacity makes the AI tech tantalizing, with the learning capacity the thing that groups such as the IEEE Computer Society predict will begin to be adopted within the next year.
The ability of a platform to perform a task requires significant amounts of data as well as defined variables. Acquiring copious amounts of data, and the right kind of data at that, is demanding, and often requires significant investment, which is why we're likely to see truly functioning cross-trained AI systems first emerge from the companies that have the capital to do it.