NEW YORK—The artificial intelligence metaphors were flying this week at Inforum 2017, ERP vendor Infor’s annual user conference.
“AI is the new steam engine, the new electricity, and it will transform everything,” said Infor President Duncan Angove. “AI is the new UI.”
Infor’s “Coleman” artificial intelligence platform, unveiled here, was named in honor of pioneering mathematician Katherine Coleman Johnson, whose work in NASA’s early space program was depicted in the movie “Hidden Figures.”
Coleman fits right in with today’s other AI brands, including Salesforce’s Einstein and IBM’s Watson. But there are similar platforms from business app providers like Oracle and dozens of other services available from cloud providers Amazon Web Services (AWS), Microsoft and Google.
AI is being talked up everywhere these days in enterprise computing, but at this point users are kicking the tires more than anything, trying to find some marginal value with AI. Meanwhile, vendors are fighting for mind share around how artificial intelligence can ignite the next wave of digital transformation. To hazard another cliché: AI in business apps right now is a me-too feature. So how do the services really stack up?
The Coleman AI is the latest to be released, but it has been in the works for several years. Infor has been a major partner of AWS since 2014, and AWS is where Infor has been building out its CloudSuite platform of enterprise resource planning (ERP) applications.
Around the time of the AWS deal, Infor formed its Dynamic Science Labs in Cambridge, Mass., right near the Massachusetts Institute of Technology. The data scientists there, led by MIT Ph.D. Ziad Nejmeldeen, are working on ERP-specific AI services to help automate processes such as “asset predictive maintenance” and “maintenance schedule optimization,” Nejmeldeen said. Others are in the works for Infor’s human capital management, retail, health care and asset management applications.
The AWS relationship is key, however, because it gives Infor an inside track to using AWS’ AI services and running them in the AWS cloud along with CloudSuite. Infor also is creating within Coleman a “conversational UX [user experience].” Voice commands using the AWS Polly speech recognition service are relayed to AWS Lex bots to carry out functions, such as managing Outlook or accessing customer information. Infor also has plans to integrate image-recognition services into applications such as asset tracking based on AWS Rekognition.
To be clear, Watson is not a talking cube going around irrigating crops, fixing elevators and spouting quotes from Hal, the ill-fated AI computer from the movie “2001: A Space Odyssey.” IBM TV ads are trying to dumb down what is in reality an extensive collection of APIs and services that can be accessed by developers in IBM’s Bluemix cloud.
The key to success with Watson is getting the APIs into the hands of developers, which is the reason for IBM’s growing Bluemix Garage project. The newest Bluemix Garage opened this week in New York (see photo). At the Bluemix Garages, startups come in to get off the ground with a new idea around a Watson service. Also, established companies such as investment firms or banks can come in to test the new Watson Financial Services to help automate portfolio analysis.
Salesforce went on an AI acquisition spree in 2013-2016, spending more than $1 billion to piece together AI solutions that are now collectively known as Einstein. The most significant purchases were BeyondCore for data discovery, MetaMind for deep learning and PredictionIO for machine learning.
Salesforce has been slowly rolling out Einstein services in its various cloud platforms over the past year. Those include such solutions as Lead Scoring (Sales Cloud), Social Insights (Marketing Cloud) and Discovery (Analytics Cloud). There are also platform services to add image recognition and sentiment analysis into any app.
Oracle last year announced Adaptive Intelligent Apps (AIA), which, similar to Einstein, can help add value to existing Oracle ERP, customer relationship management (CRM), marketing and HR applications. Oracle bases much of AIA on its Data Cloud, which is comprised of acquisitions Crosswise, Datalogix, Bluekai and AddThis, among other technologies and data providers. Together they give Oracle users access to billions of customer data points to augment analytics within Oracle ERP.
AWS, Google and Microsoft
Despite the flurry of activity among business application vendors, the AI platforms being built by the major cloud providers are key to long-term viability of AI services in the enterprise. The Big Three, plus IBM, all provide AI services such as image recognition, but more important are the engines available for data scientists and developers to use to create their own machine learning models for custom applications.
The main advantage of consuming AI services from the cloud vendors is they have the compute power to run any AI training and inference workloads, and have access to large amounts of data already in the cloud.
The final thing to remember about IT services from these and other vendors: None of this is true AI, with human-like reasoning and interaction. At best, these are helper applications that can automate and optimize existing processes. It will take time and many more breakthrough use cases for AI to start making a meaningful impact on the bottom lines of the vendors and their customers.
Scot Petersen is a technology analyst at Ziff Brothers Investments, a private investment firm. He has an extensive background in the technology field. Prior to joining Ziff Brothers, Scot was the editorial director, Business Applications & Architecture, at TechTarget. Before that, he was the director, Editorial Operations, at Ziff Davis Enterprise. While at Ziff Davis Media, he was a writer and editor at eWEEK. No investment advice is offered in his blog. All duties are disclaimed. Scot works for a private investment firm, which may at any time invest in companies whose products are discussed in this blog, and no disclosure of securities transactions will be made.