We’re halfway there.
June marks the mid-point of an already eventful year for manufacturing and industrial technology. Governmental influence on tariffs and sustainability benchmarking has reset the way many manufacturing companies are forecasting the ensuing quarters.
And, of course, rapid advances in AI technology have predictably brought it to the forefront of many industrial decision-makers’ minds. We have seen early excitement for Generative AI use cases shift to Agentic AI pilots among manufacturers over the last six months.
On November 25, 2024, Anthropic open-sourced its Model Context Protocol (MCP) standard. Since then, interest in the standard has continued to grow rapidly in both the vendor and end user community—in ways that were hardly foreseeable last December when many of us were making predictions for the year ahead.
Source: Google Trends. Numbers represent search interest relative to the highest point on the chart for the given region and time. A value of 100 is the peak popularity for the term. A value of 50 means that the term is half as popular.
According to Gartner, MCP enables “seamless integration between LLM-based applications and external data sources and tools.” It provides a standardized way for applications to discover and access contextual information, tools, and capabilities that can be used with LLM “function-calling” features. The potential for Agentic AI on the factory floor emphasizes the importance of mature data practices and a comprehensive industrial data strategy. More to come on that in a future post.
So, as the second quarter draws to a close, let’s look at five emerging technologies and trends in the industrial data space. I’ll also share some thoughts on the industries making the biggest digitalization gains.
Digital Transformation
Although the concept is certainly not new, digital transformation projects are being mainstreamed into operations and are no longer siloed in their own cost centers. This is new and seems to be good practice. Digital transformation will continue to have a significant impact on organizations in the industrial market. The two biggest influences for the success of a company in achieving its digital transformation may be soft skill building and the ability to operationalize transformation efforts. The companies we have worked with that are most successful have teams of professionals with a degree of digital fluency and a clear project charter or path to success.
According to the World Economic Forum, 70% of Industry 4.0 projects fail. Industrial companies building modern solutions that digitize their operations should look to these two influences as a path to more successful transformation projects.
Artificial Intelligence
AI is undoubtedly a rapidly growing trend. Currently, it appears to be more of a conceptual trend than a direct application for creating value on the factory floor. But the interest and attention around achieving AI use cases are clear. It will require industrial companies to provide models with contextualized, ready-to-use data.
We see major potential in the use of AI Agents at the factory level. One example is a machine that uses multiple agents, each with its own specific task. One agent monitors process variables; another looks for incoming work orders; another is responsible for tool changeovers. The ability to interact with AI agents that have in-depth “knowledge” in particular areas of the factory could help uncover operational dark spots, improve how processes flow, and simplify day-to-day monitoring of factory activity. Importantly, these agents will only be achieved through good data hygiene and an industrial data strategy.
Analytics & Machine Learning
The advanced level of data visualization capabilities reached by LLMs makes the analytics market poised for disruption. It is increasingly easy to use an LLM to generate elaborate, ready-to-consume analytics—provided the dataset has enough context. It seems extreme to believe that Agentic will kill SaaS, but it seems likely to transform it. The analytics solution providers that adapt, embrace, and embed should maintain their relevance. The others might be in trouble.
Machine Learning—now sometimes referred to as Traditional AI—is still trending, especially for use cases where algorithmic logic is necessary. Predictive asset maintenance and defect detection remain two of the most common first DataOps use cases among our customers.
Open Architectures
We see customers decoupling data from applications and virtualizing their data, enabled by technology like Apache Iceberg and Snowflake’s Zero-Copy Cloning.
We also see the market walking away from the IIoT platform model that promotes a full-stack, vertical approach to data management. Working with a single vendor to meet the demands of multiple layers in your technology stack may seem efficient, but this approach does not account for the diverse software environments that already exist within a multinational enterprise.
Based on many conversations with customers, partners, and analysts, the industrial market wants and requires open architectures that support an industrial data strategy. Technology like HighByte Intelligence Hub exemplifies this approach by offering agnostic software that meets the true definition of an Industrial DataOps solution. An Industrial DataOps solution orchestrates, observes, and governs industrial data while integrating the industrial and business systems an enterprise has already deployed. It leaves device drivers, data storage, and analytics to best-in-class software in the market because it’s in the customer’s best interests.
Energy Transition & Sustainability
Sustainability projects are being broken down into logical tasks, from cutting costs to transitioning to new energy sources. We can expect a ruthless pragmatism for sustainability initiatives under the current Trump administration. This doesn't mean sustainability programming will collapse, but it does mean a change in messaging. While the outcome may still be reduced power consumption and decarbonization, the message needs to focus on operating costs, margins, and competitive market advantage to keep these programs driving forward. Sustainability is cost-effective. Carbon accounting can (and probably will) be used as a trade tool.
And then there is AI. AI requires vast amounts of compute power and has a taxing effect on energy infrastructure. Electricity demand from AI-optimized data centers is projected to more than quadruple by 2030. This means we must consider “data yield,” which is the downstream impact software has on energy use for compute. By using an Industrial DataOps solution that delivers better, more focused data payloads from the edge, we can streamline and reduce compute processes for downstream use cases, thereby reducing digital waste.
Predictions for Industry Growth
Based on the aforementioned technologies and trends, three industrial verticals appear especially primed for growth heading into the remainder of 2025.
We expect the pharmaceutical industry to continue its recent strong growth in digitalization efforts. These manufacturers tend to focus on long-term and strategic plans, potentially due to the higher margins they are capable of achieving as a vertical. Additionally, they stand to gain exponential value from incremental innovations, whether it’s bringing a new drug to market, making batches more efficient, or conducting more experimental data collection, mimicking a full production operation within pilot projects.
We also expect the energy vertical to grow given the complicated data landscape. Even a minor productivity improvement or incremental yield can have a tremendous impact.
And finally, mining—especially rare earth mining—is poised for growth and digital transformation due to increased usage in manufactured goods and geopolitical restrictions on supply chains.
What do these three verticals have in common? Scale. Most often, they are operating and considering strategy at a global scale, as opposed to empowering disparate singular sites to control P&L and their own digital strategy. This decentralized approach, commonly seen in the contract electronics and automotive industries, creates a challenging environment for implementing an enterprise infrastructure. Conversely, a centralized strategy provides a better prepared environment for adopting transformation technology that promotes scalability.
Final Thoughts
Time can only tell, but it seems certain that further advancements in industrial technology will continue to shift these trends. Need proof? Just remember your level of awareness of MCP at the beginning of the year. Within six months, a subject known to few technologists quickly became a critical topic in the discussion of AI-enabled disruption that is transforming the industrial market.
All these technologies and trends, including and especially AI, rely on a healthy data foundation. Implementing a resilient, agile data foundation will ensure you’re prepared—no matter what the next six months may bring.
