We have all noticed the intriguing capabilities of artificial intelligence when our phones recognize our faces to authorize access or when our bank’s voice mail recognizes our voices to accept instructions or when our video software lets us add subtitles to footage simply by listening to the spoken word. Countless examples exist around us all the time.
Recently a visit to my downtown grocery store reminded me how deep AI is being infused in all interactions between companies and their customers: When I drive into the parking garage below the grocery, a camera recognizes my car. After shopping, I scan my QR code at the check-out counter, and when I leave the garage, my car is recognized again, it knows that I was at the grocery store, and my parking fee is waived.
What’s remarkable isn’t that the system can recognize my car, but that this AI technology can be used in such a low-cost project: It replaces at most a single headcount, so at high cost the economics of this project doesn’t make sense. The developer of the software must have found a cost-effective way to integrate the ability to recognize my car’s license plate into this application.
A major trend driving down the cost of implementing AI that enables applications such as the one described is the availability of machine learning (ML) models as cloud services. Software developers can apply these models in their applications at a cost an order of magnitude less than custom developing these models with help of highly paid data science specialists.
AI designed for the enterprise
If the economics of using AI in applications are so compelling, you should expect that enterprises see opportunities to deploy AI across all their processes, including operations, product development, manufacturing, supply chain, and customer engagement. We hear that exact thing from our customers, but also that for them many of the available AI services still fall short of their needs and that they are surprisingly difficult to leverage consistently across all the teams that want them.
The needs of enterprises go far beyond the needs of a simple parking management application. We see examples of hospitals that lower readmission rates by processing patient data with AI, manufacturers that inspect the products coming of their production lines with vision enabled drones, and fleet managers that predict when to do maintenance on their trucks based on Internet-of-Things sensor data. These applications improve efficiency, increase revenue, and become critical to the success of the enterprise.
Oracle is uniquely positioned to help our customers with AI for these business use cases. We have delivered industry-leading enterprise applications across enterprise resourcing planning (ERP), human resources, customer experience, supply chain, and numerous industry-specific use cases for many years. Our data and AI models pack a deep understanding of these domains, and we are making these models available as easy-to-use OCI services that can be leveraged collaboratively across all your developer and data science teams. Finally, we are doing all this with a deep commitment to open frameworks that use open source and third-party tools, and maintain full portability.
Webcast: Innovate with Business-First AI
We want to show you our portfolio of business-first AI services in a webcast hosted by Greg Pavlik, senior vice president and CTO for the Oracle Cloud Platform. Greg describes how Oracle AI makes machine learning useful, collaborative, and open, with business needs at the forefront and machine learning models that can be trained with industry-specific, application data.
I hope that you join us. You can learn more at Webcast: Innovate with Business-First AI.