Healthcare and medical research are increasingly becoming increasingly automated, connected and data-driven. Laboratories are utilising more instruments, digital systems and experimental data, while artificial intelligence (AI) is creating new opportunities across drug discovery, diagnostics, and the development of new healthcare technologies. The cost of developing a new drug is substantial before it reaches a patient. A 2025 analysis in JAMA Network Open estimated the median R&D cost of an FDA-approved medicine at $708 million, with the average reaching $1.31 billion after accounting for the cost of capital and discontinued programmes.
Therefore, the incentive for making medical research and development more efficient is considerable. However, AI is only one aspect of this transformation. To maximise the utility of these technologies, laboratories also require a digital infrastructure that can connect instruments, workflows, applications, and data. The challenge lies in improving research processes without necessitating laboratories to replace existing systems and procedures.
Moving beyond manual laboratory work
Manual processes remain part of laboratory research. Scientists may record observations in notebooks, transfer measurements between systems, prepare experiments, and document results manually. These processes can lead to practical challenges. For instance, handwritten notes may be difficult for colleagues to interpret and share; physical records may be damaged; and manual transcription may introduce errors. Additionally, information can become difficult to trace and track when distributed across various physical documents and systems.
Digitalisation can alleviate this administrative burden by capturing measurements and other experimental information electronically. Data can then be shared more easily, connected to other systems and traced throughout a workflow. The objective is to reduce manual work of scientists that takes time away from experiments, analysis and scientific decision-making. The benefits of modernising laboratories are becoming increasingly evident. A 2025 Deloitte survey of 104 biopharma R&D executives found that 53% reported higher laboratory throughput and 45% fewer human errors following modernisation. Yet, only 34% indicated that their laboratories were genuinely connected, with centralised data, integrated systems and some automation.
Digital does not always refer to being paperless. Imagine this: a leading pharmaceutical company was already using an electronic laboratory notebook, but its pre-formulation teams still relied on paper records and manual data entry. The solution came in the form of a pilot Laboratory Execution System (LES) across 10 workflows, replacing much of that work with guided digital execution and automated instrument data capture while linking existing equipment to the ELN. The lesson is clear: digitising a laboratory is not synonymous with connecting it.
The making of an intelligent laboratory
The definition of an intelligent laboratory depends on the research being performed and the systems supporting it. At its core, an intelligent laboratory employs data and connected technologies to enable scientists to work with greater context and control. A laboratory may utilise balances, sensors, pH meters, analytical devices and other instruments to collect measurements. It may also employ laboratory information management systems, electronic laboratory notebooks, spreadsheets and other applications to record and manage information. When these components operate independently, information can remain siloed across different systems. However, when connected, data can be captured automatically and made available to the workflows and applications that need it. This approach is relevant across healthcare and other research-intensive industries.
Pharmaceutical research, cosmetics development and food and beverage research may involve various products and scientific objectives, but many laboratory activities are comparable across industries. For instance, investigating how materials behave, measuring scientific properties, testing formulations or comparing the results of different experimental conditions are characteristicof research in any field. While the end-products may differ, the need for reliable measurements, structured data and documented workflows remains constant across scientific workplaces.
The engineering challenge: connecting existing systems
Modern laboratories can contain numerous instruments, software systems and processes. Connecting them requires more than adding another application. One technical challenge is whether existing systems provide suitable interfaces for integration. Some instruments and applications offer APIs and other open interfaces, while others have more limited connectivity. This distinction is crucial because a new platform can only exchange information with another system if that system provides a suitable means of communication.
Further, an organisational challenge arises because technology must address the problems that scientists and laboratory teams encounter. Understanding those issues is vital for designing an effective digital workflow. Instead of creating another isolated system, systems that are already part of the laboratory environment must be connected. For instance, Laboperator adopts an approach that works alongside existing laboratory infrastructure, connecting instruments and systems without requiring organisations to replace what is already in place.
A laboratory can continue using its existing instruments and applications while Laboperator provides an additional layer for connecting systems, organising workflows and transferring data between components. This can be implemented in different technical environments. A separate Kubernetes cluster can host Laboperator which can communicate with other systems through network connections and APIs. Those systems can run in other clusters, in various environments or on distinct premises, depending on the laboratory’s infrastructure.
The key requirement is connectivity between the systems, which means laboratories need not replace their existing technology environment to introduce a connected laboratory platform. Individual systems can continue to perform their existing functions while Laboperator facilitates their collaboration.
Automation is more than robotics
Laboratory automation is often associated with robotic arms and automated sample handling. While these technologies are important, automation can also occur at the software and data level. Some examples of automation include automatically capturing measurements from an instrument; transferring information between systems without manual transcription; and standardisingdata and automatically verifying checking experimental information. When these capabilities are combined, laboratories can minimise repetitive tasks and chances of errors while enhancing documentation. A more advanced automated workflow can connect instruments, robotics and software, thereby allowing an experiment to be executed with limited manual intervention.
The shift from automating individual tasks to coordinating entire experiments is already underway. In 2024, researchers introduced ORGANA, a robotic assistant for automated chemistry experimentation that combines large language models, visual perception and robotic systems. In one demonstration, the robotic assistant planned and executed a 19-step electrochemistry experiment in parallel, with chemists remaining in the loop.
In the user study, participants saved an average of 80.3% of their time when using the system. The demonstration illustrates how software can coordinate multiple stages of an experiment, enabling automated instruments and robotic systems to work together as part of a programmable experimental workflow. The value does not stem from adding AI to a laboratory in isolation but from creating the infrastructure that enables AI and other analytical tools to work with the right data and context.
Connected workflows can support reproducibility
Reproducibility necessitates that researchers understand how an experiment was conducted, which conditions were used and what data were generated. When such information is spread across physical notebooks, spreadsheets, instruments and various applications, reconstructing an experiment can become increasingly challenging.
Connected digital workflows can provide a consistent method for recording experimental processes and results. Workflows can be standardised, data can be captured automatically and information can be associated with the relevant experimental steps. Further, digital workflows can facilitate collaboration. Scientists can access shared workflows, review previous experiments and work from the same information rather than relying on individual notes or separate documents. Consequently, the sharing of experimental knowledge within a research organisation can be simplified. For healthcare and medical research, this is particularly important. Research teams need to be able to understand how experimental results were produced and use reliable information when making decisions about subsequent work.
Turning laboratory data into useful information
Laboratories generate vast volumes of data, but merely storing data does not automatically render them useful. Data must be reliable and available in a structure that systems can process and analyse. Manual data entry and repeated transfers between applications can introduce for errors or lead to incomplete information. Automated data capture can mitigate some of these issues. Standardisation can also facilitate easier comparison of information from different instruments and systems. Once data are connected and structured, they can support more than just the documentation of an individual experiment. Connected data can be used to analyse results, identify patterns and inform subsequent research.
As AI becomes more prevalent in pharmaceutical research, the value of this data infrastructure is likely to increase. McKinsey estimates that generative AI could generate $60 billion to $110 billion in annual economic value for the pharmaceutical and medical-product industries,driven in part by applications across drug discovery, development and productivity. This provides an essential foundation for AI. An AI system can only produce useful results when it has access to relevant data in a format it can process. Building that foundation requires laboratories to address data capture, connectivityand standardisation alongside AI development.
Why APIs and open connectivity matter
With evolving research requirements, laboratory technology continues to transform. New instruments, applications and analytical technologies are regularly introduced. Open interfaces and APIs enable these systems to communicate without requiring every component come from the same provider.
This principle is increasingly relevant as laboratories combine equipment, software and automation from different sources. Laboperator exemplifies this approach, providing a connectivity layer through which instruments can be monitored and controlled, while data can flow between devices, workflows, and other laboratory systems. For monitoring, this includes custom dashboards, remote control, threshold-based alerts and live camera feeds, with data that can be copied or exported in formats such as CSV, XLSX and JSON. Its workflow capabilities also support event-driven actions, conditional logic, scripting and RESTful APIs, allowing various pieces of laboratory infrastructure to work together without requiring them to share the same native software or originate from the same manufacturer.
The significance lies in connecting laboratory equipment as well as in enabling different machines to collaborate as part of a single experimental process. This creates a more flexible technology environment. A laboratory can introduce new tools while maintaining connections with systems already in use. For engineering teams, this is a crucial aspect of building laboratory infrastructure that can evolve. The aims is to establish connections between the different components so that instruments, software and workflows can exchange information where required. This approach can also assisthealthcare and medical research organisations in adapting their laboratory infrastructure as new technologies become available.
A new model for the laboratory is emerging
The development of AI, robotics and connected laboratory systems is creating new possibilities for scientific research. Automated laboratories are demonstrating how robotics and AI can be combined to execute experiments with reduced manual intervention. With advancement in these technologies, laboratories may increasingly use software to coordinate instruments, manage workflows and analyse experimental data. Nevertheless, the role of scientists will remain vital. Scientific research requires curiosity, judgement and the ability to determine which questions are worth investigating. Automation can handle repetitive tasks and provide researchers with better access to information, but it does not eliminate the need for human expertise.
Therefore, the future laboratory cannot solely depend on an increased availability of robots or replacement of manual processes with software. On the contrary, it will require an environment where scientists have reliable data, connected systems and digital workflows supporting their work. For healthcare, the potential impact extends beyond the laboratory itself. Better-connected research environments can facilitate the development of new medicines, diagnostic technologies and other healthcare solutions by providing researchers improved access to experimental information and more consistent processes.
Building a more connected laboratory
The transition towards intelligent laboratories begins with practical improvements. Instruments must be connected, with systems communicating effectively with each other. Workflows must be documented systematically to ensure repeatability. Data must be captured reliably, and Information must be available in a structure that can support analysis and future applications. These capabilities provide the foundation for more advanced automation and AI.
Laboperator addresses this challenge by connecting with existing laboratory infrastructure rather than requiring organisations to replace it. By integrating instruments, data and workflows, laboratories can reduce manual intervention, enhance access to information and create digital processes that can evolve as the research changes.
The intelligent laboratory is ultimately defined not by a single technology but by how effectively its technologies, data and people work together. For healthcare and medical research, that connection can help create laboratories that are more efficient, more traceable and better prepared for the technologies that will shape the next generation of scientific discovery. The next generation of laboratories may therefore focus less on replacing existing systems and more on making them work together. To see what this can look like in practice, explore how Laboperator connects existing instruments, automates workflows and assists laboratories across healthcare, scientific research, food and beverage, and cosmetics in preparing for what comes next. Editorial note: Originally published by Laboperator on Medium.