The ABR Intelligence Report: A Quantum Computing for AI Reality Check
Exploring innovations and analyzing trends in Artificial, Business, and Real-time intelligence.
In this issue:
A Quantum Computing for AI Reality Check
Cellular IoT Data Feeds Analytics Engines
IBM Empowers Digital Sovereignty
A Quantum Computing for AI Reality Check
As the compute demands of AI workloads continue to increase, there is growing interest in exploring how quantum computing could help. Currently, the most likely first use of quantum computing to accelerate AI workloads will be as an accelerator. Essentially, select parts of AI workloads would be offloaded to quantum computing processors, performing a role similar to that of a GPU.
However, as I noted in my Quarterly Report on Quantum Computing–the CTO edition (Q1 2026), the verdict is still out on whether quantum computing will be practical. Beyond technical issues, there are two other concerns. First, many pure-play quantum computing companies, like those in any field where technology is evolving and being developed with venture capital funding, may not make it financially over the long haul.
The second issue is whether quantum will be needed at all. During the time it takes to develop, deploy, and prove out quantum systems, traditional HPC approaches will continue to offer greater processing power. So, it will be a race to determine which of the two technologies (traditional HPC or quantum computing) can economically scale processing power the fastest.
Drilling Down into Quantum for AI Applications
A recent article by Quantum Insider on quantum computing use cases offered more insights into the practicality of using the technology for AI. It noted that while “headlines regularly suggest quantum computers could accelerate AI to levels far beyond current systems. The actual potential is more limited and further away than most coverage implies.”
Specifically, it noted that most AI workloads are unlikely to benefit from quantum computing. However, some types of AI workloads may indeed benefit from quantum computing. Those specific workloads include overcoming AI bottlenecks “such as optimizing neural network training, sampling from complex probability distributions in generative models, or quantum kernel methods that map data into high-dimensional feature spaces,” according to Quantum Insider.
ABR Intelligence News Analysis
Cellular IoT Data Feeds Analytics Engines
There is a growing demand for data that can be analyzed to improve operational efficiency and power enterprise and industrial agentic AI efforts. Where is that data going to come from? Cellular IoT, according to Omdia.
In its latest research on cellular IoT data, Omdia found that cellular IoT data is expected to reach 218.6 exabytes (EB) by 2035. The research revealed that the majority of cellular IoT data traffic will come from the automotive and logistics sectors. According to the report, transport and logistics are poised to become the next major sector for cellular IoT data traffic. In contrast, all other sectors combined will contribute less than 29% of the total traffic beyond 2025.
Omdia also noted that emerging trends like remote vision, which enables cameras to be added to a wide range of devices from delivery robots to industrial machinery, and agentic AI, which is driving growth in peer-to-peer machine traffic, will rely on cellular IoT data.
IBM Empowers Digital Sovereignty
IBM announced the general availability of IBM Sovereign Core, a new software platform designed to help organizations build and operate AI-ready sovereign environments and verify their control over them. The solution gives enterprises and governments an end-to-end approach to digital sovereignty.
The solution gets to the heart of modern digital sovereignty requirements, which go beyond simple data sovereignty. IBM Sovereign Core supports control over infrastructure, operations, and AI systems. To that end, the solution delivers an integrated sovereign software platform that combines control plane, identity, security, compliance, and AI execution functions within a single deployment model.