ra2 studio - Fotolia

IBM-Palantir pact speeds AI app delivery for hybrid clouds

The new Palantir IBM Cloud Pak for Data is designed to help inexperienced users create AI-based applications that can work across multiple hybrid cloud platforms.

IBM and Palantir Technologies announced a partnership this week that brings together IBM's hybrid cloud data platform with Palantir's operations platform to make it easier for users to build and deploy next-generation AI-based applications.

The new product, called the Palantir IBM Cloud Pak for Data, uses Palantir's Foundry analytics platform and will supplement IBM's Watson and Red Hat OpenShift. The new offering helps users with little or no technical training analyze huge amounts of data scattered across hybrid clouds.

IBM is targeting IBM Cloud Pak for Data at industries that typically handle large amounts of data and could most benefit from AI-based applications, including retail, financial services, healthcare and telecommunications. It will be sold through Big Blue's worldwide sales force of 25,000, compared with Palantir's 30 sales people.

Goals of the IBM, Palantir partnership

"[IBM is] not only targeting industries with large amounts of data, such as financial services and telcos, but they just rolled out their telco and financial cloud stacks," said Daniel Elman, a research manager at Nucleus Research. "They are looking to wall off areas in the cloud market that have heavy workload requirements."

For the financial services market, the product can help users with tasks involving rapid data integration, cleansing and deduplication, as well as mapping to a common data model. This capability can then be extended to use cases involving customer service and crisis management. For users in the telecommunications industry, the tool can help combine data from a range of different suppliers, CRM applications and sales orders.

While Palantir benefits from IBM's significantly larger sales force, it hasn't had access to IBM -- specifically IBM Watson. Watson users also gain an advantage by having a product that now makes the notoriously difficult Watson easier to use, according to Elman.

Watson has a number of great capabilities, but the criticism of it has always been, 'How do I use it?' or 'How do I make it deliver real-world value?' This product at least offers a potential roadmap for achieving that.
Daniel ElmanResearch manager, Nucleus Research

"Watson has a number of great capabilities, but the criticism of it has always been, 'How do I use it?' or 'How do I make it deliver real-world value?'" he said. "This product at least offers a potential roadmap for achieving that."

Red Hat will containerize IBM Cloud Pak for Data by making it compatible with OpenShift, which should improve its competitive chances, according to some analysts. It also contributes to reducing silos, integrating data across multiple hybrid cloud environments and governing data throughout the lifecycle of an AI implementation, they said.

Containerization "is a chink in the armor that needed to be filled," said Frank Dzubeck, president of Communications Network Architects, Inc. "Adopting OpenShift will allow it to run anywhere it needs to across hybrid clouds. This is a capability Palantir doesn't have."

Another benefit the deal offers to Palantir users is they can now work directly with IBM's Data Science and AI Elite Team to take on data science use cases that can address some of the challenges of AI adoption. Fiserv, a provider of financial services technology, believes the offering can give it a competitive advantage by more "securely harnessing data and creating advanced models for automating decision-making," according to Guy Chiarello, head of Fiserv's technology innovation group, in a statement.

Next Steps

IBM hybrid cloud, AI offerings ease data integration

IBM hangs its hopes on hybrid clouds and AI to revive growth

Dig Deeper on Cloud deployment and architecture

Data Center