NL logo

BLOG
Innovation Agents on a mission

Every month, an innovation agent from ChemistryNL gives you insight into his/her daily work. Read the blogs of:Martin van Dord, Caroli Buitenhuis, Anton Schotman, René Reijtenbagh, Eric van Sprang, Marco Tibaldi and Robert-Jan Lamers.

This month:Marco Tibaldi

Blog The Nobel-worthy chemistry: AI beyond academic research

This year’s Nobel Prize spotlighted the groundbreaking role of artificial intelligence (AI) in advancing chemistry and physics—a clear signal that AI has become a transformative force in science. But what are the applications of AI in the chemical industry at large?

According to Goldman Sachs, expert AI systems will gradually emerge in 2025. These “large expert models” are able to plan and execute complex, long-running tasks thanks to deep, industry-specific knowledge. For chemists, this means challenging long-held biases, by uncovering unconventional active sites on molecules, discovering novel relationships, and optimizing reaction conditions—all in ways we might not have considered. For example, an AI model suggested a new chip design, based on a “complicated electromagnetic structure… counterintuitive, unlikely to be developed by a human mind.”

Let’s explore three of the most influential AI-based technologies in chemistry today. And if you’re curious for a deeper dive, check out this comprehensive article detailing 20 technologies: link.

1. AI-Driven Drug Discovery

AI has already revolutionized drug discovery, with some AI-discovered drugs already in human clinical trials. This progress didn’t happen overnight but has come in three waves of innovation:

  • 1st wave: Mining public datasets, to identify small molecule targets and optimize candidates.
  • 2nd wave: Combining AI automation with high-throughput data generation to refine models.
  • 3rd wave: Exploiting iterative loops of highly relevant data generation combined with generative AI.

Startups are thriving in this space, with a rich database of players compiled by Anna Marie Wagner (view here). These companies are leveraging AI to accelerate timelines, reduce costs, and explore uncharted therapeutic pathways.

2. Natural Language Processing (NLP) in Chemical Space

NLP isn’t just for chatbots. By sifting through vast amounts of open-access and proprietary documents, NLP helps uncover hidden gems of information. Consider Zeta-Alpha, a startup offering a ChatGPT-style search engine for your R&D archives: it helps to “discover, organize, and reuse” knowledge. Their collaboration with BASF highlights the value of such tools in rediscovering ideas. They say in chemistry, every new idea is just rediscovering what a German chemist did in the 1950s. NLP’s ability to “think like a chemist” and extract insights from decades of data ensures that no innovation is left behind.

3. Engineering and Smart Control

AI isn’t just for labs—it’s running entire chemical plants. For example, at ENEOS Materials in Japan, an AI system autonomously operated a chemical plant for 35 days. Initially, operators were skeptical, calling to ask if it was safe to let the AI control valves in ways they hadn’t considered. Over time, trust grew as the AI demonstrated its ability to optimize operations beyond human intuition. Such advancements underscore AI’s potential to transform plant management, reduce emissions, and create safer processes.

A Glimpse Into the Future

These are just three examples of how AI is reshaping chemistry, but there’s so much more to explore. Just to spark your curiosity, a few articles from the last month:

  • IBM’s open-source foundation models promise to create a “ChatGPT for molecules materials” (read more).
  • L’Oreal will use GenAI to source all its future cosmetic formulas, to be as sustainable as possible without reducing performance. (read more)
  • Human Digital Twins: Thanks to data Integration and predictive algorithms, AI could one day replace animal testing by simulating human organs and tissues. (read more)

We’re at the beginning of this curve, so it’s too early to judge the full impact of these technologies. Instead, we should focus on enabling progress: producing better data, building robust infrastructure, and ensuring AI is powered by green energy sources.

Interested in staying updated? Follow the ChemAI LinkedIn page for the latest insights.

More information about our innovation agents:
Free expert support to accelerate
Meet our innovation agents