The newest generative AI-based assistant designed to assist with automation and simplify routine tasks is now available on the IBM mainframe platform. The Assistant for Z v1 from IBM’s watsonx combines automation coding with watsonx AI technology, making mainframe operations easier.
As stated in a blog post by Parul Mishra, IBM Digital Labor’s vice president of product management, and Skyla Loomis, IBM Z Software’s vice president, users can effortlessly import trusted automation created with Ansible, job control language (JCL), and REXX into the product catalog as skills.
Through a contextually aware chat experience, users can activate these skills. The extensible automation framework of watsonx Orchestrate enhances productivity and expedites mainframe efficiency, allowing users to build and manage AI resources and workflows.
The chat-centric generative AI Large Language Foundation model, granite 13b.labrador, powers the watsonx Assistant for Z. This enterprise-scale, English-language system is trained using a dataset of more than 2.5 trillion tokens, including IBM Z-specific content. The Z domain-specific retrieval-augmented generation (RAG) model delivers curated content to users, according to Mishra and Loomis. For instance, the assistant can produce responses to questions from IBM Z users based on Z-specific content. It can also suggest applicable automated skills for the user to apply during the conversation.
An AI assistant builder provides access to a low-code generative AI and automation studio for the erection of large language model (LLM) powered assistants, which are rooted in business context, data, and automation, according to the bloggers.
“The AI assistant builder allows you to capitalize on your automation investments and broaden adoption through a highly conversational interface loaded with trustworthy AI. Generative AI features are innately provided by IBM’s trusted and transparent LLMs, which are refined for specific business functions,” the bloggers mentioned. “Builders get the latitude and control to decide the timing and manner of LLM usage through content grounding, AI-guided actions, and conversational skills.”
In the future, the intention of IBM is to create an IBM WatsonX Assistant for Z, which operates on IBM Linux on Z. The concepts are identical for the Linux distribution, implying that patrons can simplify operations and development on IBM Linux on Z.
IBM Z is only the most recent among countless AI-based assistants provided by diverse vendors, such as Cisco, Juniper, Extreme, Microsoft, Fortinet, and many others.
A recent Gartner report stated that by 2028, 75% of enterprise software engineers will use AI code assistants, up from less than 10% in early 2023. Sixty-three percent of organizations are currently piloting, deploying or have already deployed AI code assistants.