51乱伦

Taking stock

With the quarter closed and this week鈥檚 news flow fairly subdued, it is a good moment to step back and review the big themes emerging over the past three months from all of the weekly science and innovation notes.

Across life sciences and technology, the messages are plain. AI continues to dominate. Eye-watering amounts of capital is clustering around scale players, policy is stiffening around sovereign AI, traditional sectors are retooling, and the UK government is courting global players through trade diplomacy. Infrastructure is central, with big tech companies investing huge sums in supercomputing, data centres and the energy required to power them.  Under the surface sits a persistent view that an AI investment bubble will eventually burst.

Meanwhile, the UK life sciences picture currently runs on two screens: a steady drumbeat of positive investment, ambitious government growth strategies and scientific firsts alongside headline retrenchments and pauses and a pivot towards the US, particularly in manufacturing. Signals from this week indicate the Government is listening and responding to concerns. Of course it is vital to not loose sight of the full story in all the noise.  The structural picture still matters most: patent expiries, skills shortages, AI鈥檚 impact on workflows, the risk鈥搑eward of R&D, erosion of public trust, ESG, outsourcing, open innovation, the evolving geography of innovation, global health spending, unmet need and demographics are just some of the forces at play. Real estate that aligns to these currents wins.

Beneath the national totals, regional strengths are hardening. Leeds is scaling its health tech credentials, the West Midlands is carving out a MedTech niche, and the North West is consolidating as the largest UK science cluster beyond the Golden Triangle. Apply the same lens to the wider science and tech economy and fresh seams appear in autonomous vehicles, fintech, health tech, circular economy, clean tech and defence tech.

Competition to attract science and tech firms and the talent they covet is intensifying. The UK still leads Europe and remains in the global top ten ranking of innovation clusters and universities, yet on some measures it is slipping even as others, notably China, advance. The message for those with influence is plain. Keep pace on incentives and tax, step up commercial diplomacy, market the UK offer with conviction, and resist complacency about strengths such as academia. On the flip-side rising competitor nations and regions also open the door to selective international partnerships.

For real estate the main impacts from all of this are:

Prioritise AI-driven demand. Most capital is flowing into infrastructure, yet second-order demand will rise for offices, labs, showcase centres, training facilities, AI campuses and centres of excellence. From a building specification and design perspective, prioritise digital capacity, energy resilience and robust data security. The workplace must be ready for digitised workflows and hybrid collaboration, with every setting from individual desks to meeting rooms enabling seamless interaction between people, teams and AI. Aim to be the partner of choice for AI companies and for traditional sectors undergoing major operational and identity shifts. Remember that science and tech follow hype cycles, so progress is rarely a straight line, and some players will fall away.

Track the data in detail. Maintain dashboards by city, cluster and sub-sector beneath the science and tech umbrellas to spot opportunities ahead of the pack. Track fundraising, government policy, investor intent, and company births and deaths.

Bang the drum on strengths. Make the pitch to prospective occupiers explicit, backed by data and track record if applicable: This is the right location, and your scheme will accelerate the occupier鈥檚 strategic agenda.

Align with trade diplomacy and identify international partnerships.  Track trade agreements and their outcomes, along with forthcoming initiatives at both local and national level. Stay close to inward investment agencies. Identify regions where science and technology are ascendant to cultivate partnership opportunities and prospective occupiers. Offer a soft landing for international entrants, with swing space, regulatory guidance and smooth pathways into partners and customers.

Don鈥檛 forget about the structural shifts in all the noise. Embrace the sector鈥檚 structural drivers or risk being left behind when the market turns. Map likely evolution over the short, medium and long term, and, while precise forecasting is sometimes difficult, try to scenario plan possible real estate implications.

Of course, as ever there was some strategically important news this week:

and development manager Populate announced that the Innovation Gateway incubator at the London Cancer Hub is now 85 % occupied. Two fast鈥慻rowing oncology companies (Kindling Bio and Beckman Coulter) have taken space in the 3,500 sq ft facility.

Medical鈥慽sotope supplier will open a facility at Oxford鈥檚 Churchill Hospital to manufacture positron emission tomography (PET) tracers for neuroimaging and cancer diagnosis. The plant will serve hospitals in the Midlands, Greater London and the South West, expanding Curium鈥檚 PET capacity across England.

US-based AI robotics company has received over $1 billion in committed capital in its latest funding round that valued the firm at $39 billion post money. Figure has ambitious plans to ship 100,000 robots by 2050 and this week it launched its latest humanoid robot. Morgan Stanley estimates that the humanoid robotics market could be worth $5 trillion by 2050, however full-scale adoption won鈥檛 kick in for more than a decade. Current statistics do show growth in action though. The new World Robotics 2025 statistics on industrial robots showed 542,000 robots installed in 2024, more than double the number 10 years ago. Annual installations topped 500,000 units for the fourth straight year. This means that currently, more than 4.7 million robots are in use in factories. Asia accounted for 74% of new deployments in 2024, compared with 16% in Europe and 9% in the Americas. SoftBank are getting behind the trend. This week Masayoshi Son, founder of SoftBank said 鈥淪oftBank鈥檚 next frontier is Physical AI鈥�

Investors have poured $192.7 billion into AI startups so far this year. Investors are concentrating capital into top players like Anthropic.  The US is garnering the lions share of funding, accounting for close to 80% of VC funding this year. The UK sits in second place with a 3.44% share.  Despite this noise continues related to an AI bubble about to burst. This week the Bank of England said that surging valuations of AI companies are elevating the risk of a 鈥渟udden correction鈥�.

Amid efforts to strengthen its global manufacturing network, the company  is investing more than $1 billion into its Indian operations, with plans to tap local partners for production deals.

It appears that for now at least the US government is pushing pause on the plan to raise branded pharma tariffs until its drug pricing and infrastructure spending negotiations with the industry can play out. At an event to publicise the deal with Pfizer highlighted in last weeks note Commerce Secretary Howard Lutnick said the government would let the ongoing talks "play out" as officials continue to consider pharmaceutical tariffs. 

The has drawn up proposals to increase the amount of money the NHS pays for drugs. Officials briefed the Trump administration on fresh proposals to adjust how the NHS prices medicines earlier this week, two industry figures told POLITICO. The core element of the plan includes raising the National Institute for Health and Care Excellence (NICE) threshold by 25%. The NICE threshold measures whether a treatment offers good value for money. Under the current rules, if a drug costs the NHS between £20,000 and £30,000 for every extra year of good-quality life it delivers to a patient, it is considered good value. Experts were cautious about reports that ministers may raise the price the NHS pays for branded drugs. Health economists noted that lifting NICE鈥檚 cost-effectiveness threshold would likely expand access to some innovative medicines but, given finite budgets, would also displace more cost-effective care elsewhere and could reduce overall population health. Several argued the existing threshold already sits too high once opportunity costs are properly counted, and warned that any change driven by trade or industrial policy should be transparent about who bears the health loss.

Since the launched in June, the Government reports more than £250 billion of investment, supporting an estimated 45,000 jobs across the strategy鈥檚 growth sectors. As ever, some of these commitments may have been long planned.

A recent IDC survey of 155 chief executives in the US and Canada, sponsored by Salesforce, suggests boardrooms are braced for agentic AI.  99% of respondents are prepared to integrate digital labour into their business and 73% agree that digital labour will transform their company structure.  CEOs see the first shocks in four functions. Security and software engineering will harden and speed up as code is written, tested and defended by mixed teams of humans and agents. Marketing will become a live experiment, with content and campaigns generated, shipped and measured in near real time. Operations will gain semi-autonomous schedulers, procurement helpers and field-service copilots. In short, the early wins sit at the junction of data access, repeatable rules and rapid feedback.  Crucially, the survey indicates that this is not a mass job displacement story. Four in five workers are expected to stay in role or be redeployed. 72% of CEOs predict that they and their employees will have agents reporting to them within the next five years. The imagined end state is a mixed workforce in which most employees direct at least one agent. It should be noted that to be included, respondents needed to report 鈥宲lans to adopt digital labour in the next two to five years and be knowledgeable of their organisation鈥檚 digital labour initiatives.

A future check-up could begin with a glass of water and a tiny robot. Chinese researchers have built a pill-sized capsule that can be steered by magnets to collect liquid from precise spots in the gut. In ex vivo pig-intestine tests the device oriented itself, opened a flexible internal chamber and sucked in fluid, which stayed sealed until retrieval. The prototype measures roughly 16 by 24 millimetres and needs no on board battery. It is early work, but it could lead to earlier diagnosis of conditions.

, a company founded by former OpenAI and Google DeepMind researchers, emerged from stealth last week with $300 million in seed capital from Andreessen Horowitz, Nvidia, Jeff Bezos and Eric Schmidt. Its goal is to build AI scientists,  closed-loop, self-driving laboratories where AI agents not only propose hypotheses but also run and analyse experiments. There is an initial focus on superconductors, semiconductor thermal issues, and materials R&D where experiments are verifiable and simulations can accelerate iteration. One drawback of automation is energy use. Gensler, a design firm, warns that automated labs can consume up to ten times more energy than conventional labs due to robotic work cells, dense computing and higher HVAC demands.