The Industrial AI (R)Evolution: How Agentic AI will Unleash the FULL Industrial AI Toolbox

What now seems like eons ago, in mid-2023, ARC Advisory Group coined the term "Industrial AI (R)Evolution" to capture the transformative impact artificial intelligence was poised to have on the industrial sector. While advising ARC clients, it became clear that Generative AI (GenAI)—despite its groundbreaking capabilities in the form of Large Language Models (LLMs) and the massive investment it attracted—was just one of many critical tools in the broader Industrial AI toolbox.

Consider this quote from ARC's The Industrial AI (R)Evolution Strategy Report 2023:

"Industrial software has been leveraging a variety of AI techniques and systems for decades, and generative AI adds to a portfolio that will likely expand to new systems for Causal AI, Neuro-Symbolic AI, and Quantum AI as similar breakthroughs occur in those fields of AI research."

- Colin Masson, ARC Advisory Group

This perspective was, and remains, crucial. Even before the GenAI boom, triggered by OpenAI's release of ChatGPT 3.5 in November 2022, ARC's Q4 2022 survey revealed that our industrial clients already identified AI as the most impactful technology over the next five years. Two years later, as highlighted in our recent blog post The Industrial AI (R)Evolution: Navigating the AI Wars in 2025, AI still leads our rankings of the most impactful technologies, and by a significant margin.

However, the industrial sector is often perceived as a laggard in GenAI adoption. This perception, while superficially true in terms of large-scale, public-facing deployments, lacks a critical nuance. The industrial sector has been justifiably conservative in adopting GenAI at scale, and for good reason. Many of its potential use cases for AI—think real-time process optimization, predictive maintenance, autonomous control—demand low latency, high predictability, and absolute mathematical and engineering accuracy. These are areas where the current generation of GenAI, with its inherent probabilistic nature and potential for "hallucinations," (generating outputs that are factually incorrect or nonsensical) can fall short.

2024: The Year of GenAI Contextualized

While 2024 was widely touted as the year of GenAI, for the industrial sector, it was more accurately the year of contextualized GenAI. It was the year we recognized GenAI as a powerful new tool but firmly placed it within the broader Industrial AI toolbox. This is vividly illustrated in ARC's Q4 2024 End of Year Survey for 2025, which specifically depicts a breakdown of AI technologies being applied in the industrial sector.

The Wide Range of Tools in the Industrial AI Toolbox

Here are the key insights from the ARC survey, reviewing GenAI’s impact in 2024 and anticipating trends for 2025:

• GenAI as the "New AI" (or GenUI): GenAI occupies a significant, but not dominant, position. It's recognized for its transformative potential, particularly in creating a new generation of user interfaces (GenUI) that are more intuitive and accessible.

• Knowledge Capture and Transfer—The Killer App: The survey clearly indicates that GenAI's most impactful application within the industrial sector is in knowledge capture and transfer. This is critical for addressing the long-standing and ever-growing skills gap. Experienced workers are retiring, taking decades of tacit knowledge with them. GenAI-powered systems can capture, organize, and make this knowledge readily available to newer employees, accelerating training and improving decision-making.

• The Rise of Copilots and Assistants: A flurry of vendor product announcements, and numerous customer presentations at the ARC Industry Leadership Forum in February 2025, confirm this trend. Companies are actively deploying GenAI to create domain- and enterprise-specific copilots and assistants. These tools augment human capabilities, providing real-time insights, recommendations, and assistance, always with a human in the loop to mitigate risks and ensure accuracy.

• Beyond GenAI—A Diverse Toolkit: Crucially, our survey demonstrates that most respondents recognize the need to apply specific AI tools and data science techniques for each skill and use case. This is where the broader Industrial AI toolbox comes into play.

• Causal AI, Neuro-symbolic AI, and Agentic AI: In Addition to GenAI, these also show up as major AI technologies being used:

  • Causal AI: This is crucial for understanding cause-and-effect relationships in complex industrial processes, facilitating root cause analysis and enhancing predictive capabilities.

  • Neuro-Symbolic AI: This approach integrates the strengths of neural networks for pattern recognition with symbolic reasoning for logic and rules, resulting in AI systems that are both more robust and explainable.

  • Agentic AI: This refers to systems in which multiple AI "agents," each with specialized capabilities and trained on different data or employing distinct AI techniques, collaborate to achieve a shared objective.

Agentic AI: Orchestrating the Industrial AI Orchestra

Agentic AI marks a significant advancement, shifting from isolated AI models to systems where multiple AI "agents," each with specialized capabilities, work together to achieve a common goal.

Here's a breakdown:

• What it is: Agentic AI involves creating a system of interacting agents, each trained on a specific task or dataset. These agents can communicate, negotiate, and collaborate to solve complex problems.

• Why it is Important: It allows for much greater flexibility and adaptability. Instead of relying solely on a single, monolithic AI model (like a massive LLM), you can orchestrate a team of specialized agents, each leveraging the best AI/ML technique for its specific task. This includes not just GenAI agents, but also agents built using machine learning, deep learning, causal inference, and other techniques from the Industrial AI toolbox.

• Not a Universal Solution: Just as GenAI isn't the answer to every Industrial AI use case, Agentic AI isn't needed for every situation. Simple, well-defined tasks might be better served by a single, specialized model.

• Relieving the Pressure: Agentic AI is expected to significantly relieve the pressure on industrial-grade data scientists and AI architects. Instead of forcing every problem into a GenAI framework, they can now focus on designing and orchestrating highly specific AI agents, leveraging the full range of tools in their Industrial AI toolbox. This enables more efficient resource utilization and enhances overall performance.

The War for Talent: Why Engage with ARC

The Industrial AI (R)Evolution is not just about technology; it's about people and processes. It’s about AI talent. The ability to attract, retain, and develop individuals with the skills to build, deploy, and maintain these sophisticated AI systems is paramount. This is where the war for Industrial-grade AI talent is being fought.

The most important reason for engaging with ARC Advisory Group analysts and attending ARC Forum events, including the Industry Leadership Forum in Feb 2025, and upcoming Industrial AI Leadership Summits, is to tap into our deep domain expertise. We provide unparalleled guidance on:

• Identifying the Right Technologies: Understanding which AI tools and techniques are best suited for specific industrial use cases.

• Vendor Landscape Navigation: Gaining insights into which vendors are leading the way in Industrial AI, and which ones possess the strongest talent pools.

• Best Practices and Strategies: Learning from the successes (and failures) of other industrial companies on their AI journey.

• Talent Acquisition and Development: Discovering strategies for building and nurturing the AI expertise needed to succeed in this rapidly evolving landscape.

The Industrial AI (R)Evolution is underway. It's a journey that requires a pragmatic, multi-faceted approach, and ARC Advisory Group is here to guide you every step of the way.

For ARC Advisory Group recommendations for navigating the AI Wars, closing the digital divide by embracing Industrial AI, and governing and guiding major decisions about enterprise, cloud, industrial edge, and AI software, please contact Colin Masson at [email protected] and set up a meeting with me, or my colleagues, at the ARC Leadership Forum!