Key Takeaways
- AI is accelerating materials simulation and chemical R&D, allowing researchers to evaluate far more candidate materials in less time.
- In manufacturing, predictive maintenance is emerging as a practical AI use case because companies can use historical operating data to test and validate models.
- AI-powered customer support and knowledge retrieval can make complex chemical product information easier for customers to access.
- AI could reshape chemical sales by lowering the cost of serving smaller customers and enabling more self-service product selection and formulation.
- As companies automate more work, they also need to protect the practical, experience-based knowledge held by scientists, operators, salespeople, and other employees.
Artificial intelligence is beginning to change how chemical companies discover materials, operate plants, support customers, and sell products.
But AI’s impact on the chemicals industry will not happen in the same way, or at the same speed, across every function.
In Lux Research’s webinar, “How AI Will Transform the Chemical Enterprise Over the Next Decade,” four areas stand out as particularly important: R&D, manufacturing, customer support, and sales.
Here are four ways AI is changing the chemicals industry and what they could mean for chemical companies over the next decade.
1. AI Is Accelerating Chemical R&D and Materials Discovery
One of AI’s most visible applications in the chemicals industry is materials research.
Traditional computational methods can require significant time and computing resources to simulate molecular and material behavior. AI-based approaches can dramatically shorten that process.
Preferred Computational Chemistry’s Matlantis platform, for example, uses a neural network trained on approximately 60 million density functional theory simulation results. This allows researchers to perform certain calculations much faster than with conventional DFT methods.
AI Can Dramatically Reduce Materials Simulation Time
| System Size | Conventional DFT | Matlantis |
|---|---|---|
| 256 atoms | 2 hours | 0.1 seconds |
| 3,000 atoms | ~2 months* | 0.3 seconds |
Faster simulation can allow researchers to evaluate much larger materials design spaces and eliminate poor candidates earlier in the development process.
Generative AI can also propose new molecules and materials, while greater laboratory automation could make it possible to test more of those candidates.
However, faster discovery does not automatically translate into faster commercialization.
A promising new material still needs to be synthesized, characterized, tested for a particular application, manufactured at scale, and produced at an economically viable cost.
For chemical companies, the real opportunity is therefore broader than simply discovering more materials.
AI needs to improve the entire journey from initial material discovery to a commercially successful product.
2. AI Is Improving Chemical Manufacturing and Predictive Maintenance
AI is also creating opportunities inside chemical manufacturing operations.
Predictive maintenance is one of the more mature applications because chemical plants already generate large amounts of operating and sensor data.
Instead of relying only on predetermined maintenance intervals or waiting until equipment begins to fail, AI models can analyze multiple operating signals and identify patterns that may indicate developing problems.
C3.ai developed process-specific models that combined different sensor signals to detect potential maintenance problems. When the models were tested against historical data, the analysis indicated that the system could have identified the furnace failures in advance.
This highlights an important factor in industrial AI adoption: the strongest initial applications often have a clearly defined problem, accessible historical data, and a way to determine whether the model would have produced a better result.
AI can also enhance existing manufacturing and planning systems without replacing them completely.
For example, BASF partnered with Google to use the AlphaEvolve coding agent to explore alternative modeling and forecasting approaches based on existing rules-based methods.
Rather than asking AI to independently manage an entire supply chain, the technology can augment established planning and programming capabilities.
For chemical companies, this incremental approach may offer a more practical path to deploying AI in complex manufacturing environments.
3. AI Is Transforming Chemical Customer Support
Some of AI’s most immediate applications may be found outside the laboratory and plant.
Chemical companies manage large product portfolios, safety documentation, technical specifications, product codes, and other information that can be difficult for customers to navigate.
That makes customer support and knowledge retrieval strong candidates for AI.
Covestro, for example, partnered with LoyJoy to create an AI agent that helps customers locate safety data sheets.
Instead of requiring customers to understand exactly how Covestro organizes its products and documentation, an AI interface can help users find the information they need more naturally.
AI agents can interpret customer requests and interact with established systems to support tasks such as providing order information or handling straightforward reorders.
The underlying processes already exist. AI makes them easier for customers to access.
This distinction is important.
Companies do not necessarily need to begin with highly autonomous or technically ambitious applications. One of the fastest ways to generate value can be to identify repetitive customer interactions where the correct information already exists but is difficult or time-consuming to retrieve.
AI-powered interfaces can make that knowledge available without requiring customers to navigate complex internal systems.
4. AI Is Changing How Chemical Companies Sell
AI could ultimately have its biggest commercial impact on the relationship between chemical companies and their customers.
Traditionally, selling specialty chemicals often requires significant technical support.
Customers may need help identifying the right material, understanding product properties, selecting ingredients, developing formulations, or troubleshooting an application.
AI can begin to automate portions of that process.
Evonik’s Coatino platform demonstrates how this could work. The customer-facing materials informatics platform helps users identify coating ingredients such as defoamers and dispersants. Adding an LLM interface makes it easier for customers to interact with the platform and access specialized product information.
Dow provides another example. Its materials informatics capabilities can generate plausible foam formulations designed around particular target properties, helping reduce the time required for formulation and validation.
These technologies could change the economics of chemical sales.
Historically, it may not have been economical for a large chemical company to devote substantial sales and technical resources to very small customers.
AI-powered product selection, technical support, formulation tools, and self-service platforms reduce the human effort required for each interaction.
As a result, chemical companies could potentially serve a much longer tail of smaller customers and niche applications.
There is also a secondary benefit.
Every digital customer interaction can generate information about what customers are searching for, which properties matter to them, and what technical problems they are trying to solve.
Over time, that data could inform product development, R&D, portfolio management, forecasting, and market strategy.
How Should Chemical Companies Prioritize AI Investments?
The most technologically advanced AI application is not necessarily the best place to start.
Lux Research’s AI Application Selection Framework evaluates opportunities through two broad dimensions: value and prioritization.
Value considers both the knowledge generated by an application and the value created through automation.
Companies can consider factors such as how important a task is to organizational performance, how frequently it occurs, how standardized it is, how labor-intensive it is, and how time-sensitive the work may be.
Prioritization then accounts for factors such as time to value and implementation risk.
Viewed through this framework, customer-facing applications can be particularly attractive.
Customer support, document retrieval, and materials search can create meaningful value while often being easier to implement than fully autonomous laboratories or more sophisticated AI-driven manufacturing systems.
The question companies should ask is not simply, “Where can we use AI?”
It is, “Where can AI change the economics or effectiveness of an important business process?”
The Hidden Risk of AI: Losing Human Knowledge
There is also a less obvious risk associated with AI adoption.
The webinar distinguishes between two forms of knowledge: “techne,” or systematic knowledge based on established principles, and “metis,” the practical knowledge that develops through experience.
Chemical companies are generally good at documenting techne.
It appears in operating procedures, specifications, databases, research reports, and technical documentation.
Metis is much harder to capture.
It is the plant operator who recognizes that an unusual sound indicates a developing equipment problem.
It is the scientist who knows that a theoretically promising formulation will be difficult to manufacture.
It is the salesperson who understands why a customer is reluctant to switch products, even when the technical data suggests that they should.
If a company automates the tasks through which employees develop this type of expertise, it may also remove the process by which that knowledge is created.
That means AI implementation should also be treated as a knowledge-management challenge.
Before automating a task, chemical companies should understand not only what work is being replaced, but also what human knowledge may be embedded in that work.
Frequently Asked Questions About AI in the Chemicals Industry
How is AI being used in the chemicals industry?
Chemical companies are using AI across R&D, manufacturing, customer support, and sales. Applications include materials simulation and discovery, predictive maintenance, production planning, technical document retrieval, automated customer service, materials selection, and product formulation.
Where could AI have the biggest near-term impact in chemicals?
Customer-facing applications may offer some of the strongest near-term opportunities. AI-powered customer support, materials search, document retrieval, and self-service tools can often be implemented more quickly than advanced R&D or manufacturing applications while directly improving how customers interact with chemical suppliers.
What is a major risk of AI adoption for chemical companies?
One important risk is losing practical knowledge held by experienced employees. When companies automate tasks, they need to identify what tacit expertise exists within the current process and determine how that knowledge can be preserved, transferred, or recreated.
What AI Means for the Future of the Chemicals Industry
Over the next decade, AI is likely to influence nearly every part of the chemical enterprise, but different applications will mature at different speeds.
Materials simulation and discovery could help researchers explore larger design spaces. Predictive maintenance can help manufacturers identify potential failures earlier. AI-powered customer service can make technical information easier to access. And customer-facing materials informatics tools could change how companies sell products and serve smaller markets.
The larger opportunity, however, goes beyond automation. AI can help chemical companies make specialized knowledge more accessible, lower the cost of serving customers, identify new commercial opportunities, and better understand what the market needs.
The challenge will be determining where AI creates genuine business value while preserving the human expertise that remains essential to the chemical industry.
Watch the Full Webinar
Want to explore how these trends could reshape chemical R&D, manufacturing, customer support, and sales?
Watch Lux Research’s webinar, How AI Will Transform the Chemical Enterprise Over the Next Decade, to learn where AI could create the greatest opportunities for chemical companies and how organizations can prioritize applications for the decade ahead.
