Tech giant IBM (IBM) is rolling out a new type of storage system called content-aware storage (CAS). Instead of storing data in one place and processing it somewhere else, CAS does both in the same system. For example, it can turn documents into AI-friendly formats using large language models, which is something that normally happens outside storage. Because of this, companies can run AI tasks like retrieval-augmented generation (RAG) faster, more securely, and without needing extra infrastructure.
Claim 30% Off TipRanks
Trade IBM with leverageAt the center of this system is something called a vector database, which helps AI find information based on meaning rather than exact words. Working with Samsung (SSNLF) and Nvidia (NVDA), IBM has already shown that it can handle 100 billion of these data points on a single server while keeping accuracy above 90% and response times under 700 milliseconds. This matters because RAG is quickly becoming the main way companies use AI, as it lets models combine user questions with internal company data to give more accurate answers without needing expensive retraining.
However, handling this much data is not easy, since even one file can turn into hundreds of data points, and quickly add up to billions. To solve this, IBM redesigned these databases to make them more efficient and reduce the number of servers needed. As a result, tasks that used to take months can now take days. Looking ahead, IBM and Nvidia plan to make this even faster by eventually processing 100 billion data points in a single day.
Is IBM a Buy, Sell, or Hold?
Overall, analysts have a Moderate Buy consensus rating on IBM stock based on 11 Buys, six Holds, and zero Sells assigned in the past three months, as indicated by the graphic below. Furthermore, the average IBM price target of $324.19 per share implies 36.7% upside potential.
