AI-Driven Scientific Breakthroughs: Redefining Antibiotics and the Laboratories That Power Them

In a remarkable leap for medical science, Massachusetts Institute of Technology Researchers have developed a generative AI tool that can design entirely new antibiotics against drug-resistant superbugs – an approach that has already yielded two promising candidates. Although years of refinement and clinical trials lie ahead, the development could signal a “second golden age” in antibiotic discovery, creating a new dynamic for real estate needs.

Until now, AI has typically been used to screen existing chemical libraries for potential antibiotic candidates. In contrast, the MIT team’s generative AI approach – trained on 36 million compounds alongside bacterial growth data – builds new molecules from scratch. By studying the molecular structures of atoms, such as carbon, oxygen, hydrogen and nitrogen, the system learned to predict antibacterial effects. With the clinical trial participants proving highly potent against the superbugs, experts suggest the tool allows for “an expanded arsenal” against evolving superbugs.

While demand for traditional laboratories remains strong – supporting core wet-lab functions such as biological testing and chemical analysis that continue to underpin major scientific advances – the landscape is evolving. With the acceleration of AI, occupiers’ real estate needs are shifting as quickly as the science itself. Life sciences facilities, once centred almost exclusively on bench-based research, are now being reshaped by the rise of “tough tech”. This new wave of innovation is fuelling demand for hybrid spaces that combine advanced research and development with small-scale manufacturing, pilot production and complex engineering. Such facilities must be built to accommodate heavier physical loads, enhanced ventilation and highly specialised clean rooms – not only for biological research, but also for electronics, nanofabrication and chemical synthesis, including the kind required for AI-driven discoveries.

Power requirements are rising sharply. Traditional wet labs have relatively modest energy demands, but the next generation of facilities must accommodate high-performance computing clusters, AI training infrastructure, and in some cases quantum simulation hardware – each with substantial cooling needs. Pilot manufacturing for biopharmaceuticals, advanced materials or precision-engineered components can require continuous operation of energy-intensive machinery, driving the need for upgraded electrical capacity, resilient back-up systems and robust grid connectivity.

As a result, property developers are rethinking design from the ground up. Flexible layouts must allow for rapid reconfiguration between research, testing and manufacturing, while structural load capacities need to be higher to support heavier equipment. Sites may require expanded utility corridors, advanced waste management for biological and chemical by-products and secure storage for hazardous materials. In many cases, facilities are now being planned with direct links to renewable energy sources or on-site generation to manage both cost and sustainability pressures.

This evolution is changing location strategies too. While established life sciences clusters like Cambridge and Oxford have long been central to the sector, hubs are forming in areas with more available industrial land and better access to power infrastructure. Proximity to transport links for moving prototype equipment or trial-scale production runs is also becoming a competitive advantage.

Ultimately, AI’s role in antibiotic innovation illustrates a deeper transformation: scientific progress increasingly depends on physical spaces that can keep pace with technological complexity. In a world where laboratory research is merging with engineering, computing and manufacturing, tomorrow’s breakthroughs will be forged not just in code or under a microscope, but in high-specification, power-intensive environments purpose-built to bridge discovery and delivery.