CBRE’s Life Sciences (LS) Sector recently hosted a FOCUS Forum on “Disruptors in the Lab.” 15 clients and CBRE colleagues collaborated during the virtual event which highlighted expert insights from various organizations and brought together innovators who are reshaping the laboratory landscape. The discussion explored how new AI and automation technologies may drive the future of laboratories and shape scientific research and discovery. The event also featured an exercise led by Mike Jackson, founding member from Shaping Tomorrow Consultancy, involving an AI-driven glimpse into the future of labs.

Todd Richardson, Global Head of Full Spectrum Lab Services from CBRE, introduced the panelists including Susan Crosland (Scientific Equipment Manager, Syngenta), Katy Rhynard (Associate Director R&D Information Systems, Gilead), David Bendet (Global Lab Design Leader, CBRE Life Sciences), Paul Janssenswillen (Head of Scientific Projects, Full Spectrum Lab Services), and AI expert, Mike Jackson. The discussions are summarized below, and include the AI predictions.

The Impact of AI across Life Sciences and Laboratories

The conversation started with a basic understanding of Artificial Intelligence (AI), Automation and Cloud Labs:

• AI focuses on creating machines or software that can perform tasks that typically require human intelligence. These tasks may include learning, problem-solving, understanding natural language, recognizing patterns, and making decisions. AI algorithms can analyze large amounts of data and adapt their behavior based on the patterns they discover.
• Automation on the other hand, does not use human intelligence, but rather uses technology and computer systems to perform tasks or processes without human intervention. Automation aims to streamline workflows, increase productivity, reduce errors, and save time and resources by replacing manual labor with machines, software, or robots.
• Cloud Labs are virtual laboratories that operate in a cloud computing infrastructure. In a cloud lab, scientific experiments, research processes, data analysis, and other laboratory activities are conducted using cloud-based resources, software, and platforms. This allows researchers and scientists to remotely access and manage experiments, data, and computational resources without the need for physical infrastructure.

The three technologies (AI, Automation and Cloud Lab) can complement each other, as AI-driven insights can inform automation strategies within a cloud lab, resulting in more efficient and productive research workflows.

Data from our live polling indicated that AI’s impact across Life Sciences and Laboratories will generate more automated systems in the laboratory than has been seen in the past. Automation companies are already significantly reducing human touch points to increase efficiency and efficacy. One of our panelists discussed how manual intervention brings about mistakes so reliability and reproducibility should be enhanced with automation as a natural result of using AI. This does spark the question though – does AI allow for creativity and innovation? Sometimes science innovation is thinking outside of a pattern and that is hard to reproduce outside of a machine in the real world. AI allows humans to spend more time on creativity, therefore allowing space for the lab to be automated so you can accomplish more with the same amount of space.
We also polled our participants to ask if there will be more or less product volume as a result of AI.  Data from our live poll indicates that there will be more personalized medicine product volume as a result of AI.

Automation and Space Requirements

Data from our live polling indicated that space requirements in the lab will increase due to automation. Various points were made by our panelists about how robotic systems can sometimes take up more space; but, then again can replace multiple scientists or staff doing work in the lab.  Will jobs be won or lost in this process? AI predicts that certain types of workers are likely at risk to lose their jobs like data entry clerks, lab technicians and assistants. It is also predicted that apprentices will rise stronger than seen in the past – especially genetic counselors, biomedical engineers, and bioinformatic scientists. Mixed reality is also playing a huge role in lab services as we see shifts in working environments and staffing.

What’s the impact of AI across Life Sciences and Laboratories?

The ratio of lab and office is shifting. The gold standard was always approximately 60% lab and 40% office space; in a post-COVID world we have seen a large shift in office space being under-utilized. A number of organizations are looking to increase their lab space and maximize their real estate value by having more bench space. The ratios are starting to trend to 70% lab and 30% office space, but with impacts from AI and automation we could see a continued trend where lab space could significantly exceed adjacent office space by 9X. The highest and best use of office and lab space can be attributed to collaboration around the scientific environment. AI generated insights indicate a growing trend towards more flexibility and modularity in lab design.  This might mean using empty spaces for new offices where remote admin teams can work, and adding amenities to retain talent. Some redundant lab infrastructure may also be released to optimize costs.  However, dedicated areas will continue to be essential for specialized equipment requiring controlled environments.

Will Laboratories Become More Centralized or Decentralized as a Result of AI and Automation?

Data from our live polling indicated that there is about a 50/50 split between our clients predicting that labs will become more centralized or decentralized as a result of AI, but that answer seems to be rooted in discovery vs. development where the dynamics differ. Most innovation today is happening in smaller groups. Larger organizations develop new products as well but tend to focus more around developing and commercializing acquired products. AI may make lab innovation and the data engines to fuel innovation more accessible to smaller groups and drive even greater decentralization of early stage innovation. At the same time, automation can also drive centralization of the development and regulatory approval phases of products. It also may be a challenge for organizations to try and navigate the complexity of centralizing their labs or outsourcing to a centralized CDMO solution.

Will laboratories become more centralized or decentralized as a result of AI and Automation?

We also polled our participants to ask if labs will become more centralized or decentralized as a result of automation. Data from our live poll indicates that labs will become more centralized as a result. The two go hand in hand: if AI is helping to automate then you can have a more of a decentralized lab. If your automation requires more engineers to keep it running, you may need more of a centralized lab space. The cost of automated machines requires specialized and sophisticated talent to keep these automated machines running.

Interestingly the AI predictions suggested a hybrid model would emerge – balancing centralized specialized sites and decentralized standard workflows, and this would be enabled by connectivity, sensors, fail-safes and oversight protocols.

AI and the Future

Most organizations want to see an increase in high quality science for less cost and that they want to solve problems quicker. It can be said that automation and AI help to address this. AI could potentially create greater accessibility to strong problem-solving tools in the hands of individuals, entrepreneurs, and small groups.  AI also can create large centralized science processing centers, which could proliferate with the right automation. It seems as if AI is likely going to further put the power of big data in the hands of smaller teams, and a lot of them will work and live where they chose vs. where the company decides.

“Looking ahead, standard testing workflows and commodity capabilities are expected to become more distributed or decentralized through automation, modular setups, and teleoperations.  This allows location flexibility for routine processes based on demand patterns.  However, specialized research and complex processes involving niche equipment, sentiment samples or requiring high security will remain largely centralized in main hub locations.”

AI’s response to the future of automation suggests a significant rise in lab automation and the integration of technologies like AI, ML, robotics, and IoT over the next decade. These projections are based on successful pilot programs demonstrating productivity and cost-effectiveness:

• ‘Approximately 60-80% of current routine processes—such as sample intake, barcode scanning, inventory documentation, basic sample preparation, equipment maintenance, and results archiving—are poised for increased automation through robotics.
• Processes like workflow coordination, inventory management, sample tracking, and hazard detection are also expected to become highly automated using sensors, connectivity, and algorithms.
• Tasks involving specialized sample handling, complex protocols, intuition-driven activities, final decision-making, and oversight will continue to require human specialists, at least in the medium term. Creative endeavors and abstract reasoning will similarly rely on human ingenuity, though automation will provide additional support.
• While a fully autonomous lab may not be imminent due to technical and regulatory challenges, automation is set to transform workflows, productivity, skill profiles, and the overall footprint of most standard lab processes. Guiding policy measures will play a central role in shaping and directing these transformative shifts.’