Daniel Petras, Assistant Professor, University of California, Riverside, USA
I assume generative AI will turn our lives and science upside down. The way we can analyze and integrate ultra-complex data layers (e.g., from multi-omics) has already been fundamentally disrupted by current LLMs, and I can hardly fathom what the capabilities will be five years from now. I assume we will integrate AI tools into all aspects of analytical sciences, end-to-end, from instrument development and experimental design to reporting of results and contextualization of other studies.
Tao Chen, Senior Principal Scientist, Genentech, Inc., USA
Looking ahead, the defining trend will be the seamless convergence of advanced analytical technologies and artificial intelligence (AI). As therapeutic modalities become increasingly diverse and complex, passive post-experiment data collection is no longer enough. Instead, analytical science will be pushed to generate high-throughput, high-resolution, multi-attribute data to continuously train and optimize predictive AI models. This shift enables true "lab-in-the-loop" workflows where analytical insights directly guide synthetic routes, formulation strategies, and method development across technical development in real time to accelerate the delivery of safe and effective therapies to patients.
Bob Pirok, Associate Professor, University of Amsterdam, the Netherlands
I think the greatest impact will come from autonomous and self-learning analytical laboratories. Not automation in the narrow sense of replacing manual labor, but systems that can reason across previous experiments, instruments, datasets, and objectives. The real transformation will be the creation of analytical environments with a kind of shared memory: systems that know what has been tried before, what worked, what failed, under which conditions, and for which analytical purpose.
This could change method development profoundly. Today, much expertise remains implicit in the minds of individual researchers or hidden in notebooks, instrument methods, and scattered datasets. If we can capture and structure that knowledge, future analytical workflows can become more reproducible, transferable, and efficient. The promise is not that science becomes push-button, but that researchers spend less time rediscovering what is already known and more time asking better questions.
Christopher P. Rüger, Research Group Leader, High-Resolution Mass Spectrometry, University of Rostock, Germany
Seeing the tremendous impact of artificial intelligence on our daily lives, AI may be the only valid answer to this honestly very difficult question.
With increasingly capable AI systems on the horizon, I believe the way we perform science in general, and analytical science specifically, will fundamentally change. AI is already helping researchers navigate the enormous volume of scientific literature, develop and structure ideas, write code, and interrogate increasingly complex datasets. More importantly, we are beginning to see systems that connect AI reasoning directly with experimental planning and execution. Autonomous laboratories can already operate in closed loops in which experimental results inform what experiment should be performed next.
From the perspective of mass spectrometry, I see three major areas where AI could have the largest impact: technology development, including instrument design and construction, experimental design and instrument operation, and data analysis and interpretation.
Today, we usually decide in advance what the instrument should do. We select the ionization conditions, acquisition parameters, fragmentation energies, or separation settings and then evaluate the results afterwards. Some aspects of data-dependent acquisition already bring this idea of an intelligent, real-time response to life. I can imagine that this will be a big part of many future devices and methodologies. An intelligent instrument could recognize something unusual during the experiment and adapt its next measurement accordingly. When analyzing something as complex as a pyrolysis oil or a carbonaceous aerosol, it could identify an interesting compositional region, change the ionization conditions, acquire additional fragmentation data, or explore it using another analytical dimension before we even ask it to.
Ideally, AI should give us more freedom to explore ideas, perform experiments that would otherwise be too complex or time-consuming, and perhaps even challenge us with observations we might have overlooked. Ultimately, no one knows exactly where AI will take analytical science. I hope, however, that AI will remain a tool that expands our curiosity rather than replacing it.
Henry Day, Senior Staff Scientist, Illumina, Cambridge, UK
Many careers are having to adapt to the rapid development and adoption of AI, and analytical science is no different. With increasing automation, our ability to generate vast amounts of data from analytical tools has never been greater and AI will become an increasingly powerful partner in processing that data to draw out meaningful information and accelerate discovery. As analytical scientists we will need to evolve too, focusing our expertise on designing robust experiments, asking the right questions, and applying scientific judgment to interpret and challenge the insights that AI delivers.
Martina Catani, Associate Professor, Department of Chemical, Pharmaceutical and Agricultural Sciences, University of Ferrara, Italy
Looking five to ten years ahead, I think that one of the trends that may have the greatest impact on analytical science is the integration of advanced AI models into the interpretation of highly complex analytical data. As high-resolution mass spectrometry, multi-omics platforms and hybrid separation techniques continue to evolve, the datasets they generate are becoming richer but also far more challenging to decode. AI-assisted interpretation will play a crucial role in resolving challenging spectral patterns, predicting molecular formulas for unknown compounds and identifying subtle features that are difficult to capture with traditional algorithms. I expect that rather than replacing experimental expertise, these tools will enhance it, helping scientists extract deeper meaning from their data and accelerating the path from measurement to understanding.
Francesca Rigano, Assistant Professor, Messina Institute of Technology, Department of Chemical, Biological, Pharmaceutical and Environmental Sciences, University of Messina, Italy
In my opinion, over the next five to ten years, the greatest impact on analytical science will come from the integration of artificial intelligence, machine learning, and predictive chemometric tools. In my own research, I have explored quantitative structure–retention relationship (QSRR) models to predict the retention behavior of unknown compounds, improving identification strategies in liquid chromatography. In the future, predictive approaches will increasingly support analytical method development, reduce experimental workload, and make analytical science faster, greener, and more reliable. Importantly, the role of the analytical chemist will become more critical than ever, to correctly interpret results, validate models, and ensure that analytical methods are scientifically sound and fit for purpose.
Josias de Oliveira Merib, Associate Professor, Federal University of Health Sciences of Porto Alegre, Brazil
Over the next five to 10 years, I believe the integration of artificial intelligence with automated experimental workflows will have a major impact on analytical science. AI-assisted platforms integrated online with analytical instruments could optimize extraction conditions, reduce human intervention, improve reproducibility, and enable faster, more efficient, and independent analytical workflows.
Flavio Antonio Franchina, Associate Professor, University of Ferrara, Italy
Easy… generative and machine-learning tools. Specifically, their integration into analytical workflows and instrumentation. Instruments that design and refine their own methods, and models that propose identifications, will push forward what is practically achievable. My one concern is a generation of scientists who can run an optimization but cannot explain
Newsletters
Receive the latest analytical science news, personalities, education, and career development – weekly to your inbox.
