The Wellcome Genome Campus has created a Science & Technology Advisory Group to help guide the scientific and technological direction of its expanding site.

Chaired by Dr Nicole Mather of IBM Consulting, the group includes representatives from Novartis, Genentech, the Wellcome Sanger Institute, EMBL-EBI, the Health Data Research Service and Wellcome, alongside campus leadership.

It has been asked to develop a strategy for what the campus describes as an AI-native district focused on omics, biodata and the use of AI in health and care translation. The group will advise on scientific priorities, emerging opportunities, infrastructure needs and the offer to incoming occupiers.

The move comes as the Wellcome Genome Campus undertakes a major expansion near Cambridge. The site is due to grow from 125 acres to 440 acres, with plans to support a community of 9,000 people or more and attract about 250 companies involved in research, translational development and other commercial activity.

Existing institutions on the campus include the Wellcome Sanger Institute and EMBL’s European Bioinformatics Institute, while the Health Data Research Service is also joining the site. Together, they give the campus an unusual concentration of genomics, biodata and health data organisations within the UK life sciences sector.

Members of the new group include Dr Andy Richards, Dr Avi Spier of Novartis, Professor Ewan Birney of EMBL-EBI, Dr John Marioni of Genentech Research and Early Development, Professor Matt Hurles of the Wellcome Sanger Institute, Dr Melanie Ivarsson of the Health Data Research Service, Neelam Patel and Professor Rachel McKendry of Wellcome.

They are joined by Robert Evans, Chief Executive Officer of Wellcome Genome Campus, Phil Clark, Chair of Wellcome Genome Campus, and Robert Holl of the Wellcome Trust’s investment division.

Expansion plans

The expansion forms part of what the campus describes as one of the UK’s largest recent investments in life sciences and technology infrastructure. The first phase is under way and is expected to be completed by 2028.

Plans for the wider site include new research and translation laboratories, commercial space, incubator and accelerator facilities, homes and community amenities. The project also includes additional data centre computing capacity, energy infrastructure and public realm works.

The expansion reflects a broader push across the Cambridge cluster to add laboratory space and specialised facilities as demand grows from biotechnology, data science and healthcare companies. By bringing research institutes, commercial tenants and national data infrastructure together on one site, the campus is seeking to strengthen its role in that ecosystem.

The arrival of the Health Data Research Service adds another element to that model. It is intended to provide national health data infrastructure, linking the campus’s existing strength in genomics and biological data with clinical and patient outcome information.

Advisory role

The Science & Technology Advisory Group will meet regularly in an advisory capacity. Members will also act as ambassadors for the campus in the UK and overseas.

Its creation signals an effort to shape the scientific identity of the enlarged site before much of the physical build-out is complete. It also brings in senior figures from pharmaceutical companies, research organisations and health data bodies at a time when AI and data-led approaches are becoming more central to biomedical research and clinical development.

Robert Evans outlined the rationale for the new body in a statement on the expansion strategy.

“Our expansion is about creating the right conditions for our science and technology community to thrive. The Science & Technology Advisory Group is helping us to set a clear direction for our future, foster collaboration, commerciality, talent attraction and retention and ensure we continue to grow as a world-class destination,” said Evans.

Dr Mather said the campus had an opportunity to define how biological data and AI are brought together in health research and application.

“The Wellcome Genome Campus is uniquely positioned to shape the future of a key frontier field: how we combine omics, data and AI to transform health and care for people globally. The Science & Technology Advisory Group will create a strategy that builds on the Campus’ pedigree and strengths to grasp the many opportunities now emerging,” said Mather.

e2e-assure has launched Cumulo, a sovereign AI-driven security operations centre platform for IT and operational technology environments. It describes the product as the UK’s only sovereign zero-day SOC platform.

The launch is aimed at organisations that want to keep cyber defence data and AI models within UK-controlled infrastructure, particularly operators of critical national infrastructure and businesses in regulated sectors. Cumulo combines threat detection, incident analysis and environment modelling in a system designed to reduce reliance on external cloud AI services.

The platform is built around a digital twin of each customer environment, maintained continuously through passive discovery across IT and OT systems. That replica is used for attack simulation and to identify risks before they are exploited, an approach e2e-assure says is particularly relevant in operational environments where live testing can disrupt services.

Another element is what e2e-assure calls a zero-day SOC model. In practice, this means live threat intelligence can be turned into detection rules immediately rather than waiting for slower update cycles, with the aim of narrowing the gap between a threat emerging and an organisation being able to detect it.

The system keeps artificial intelligence alongside a security information and event management platform rather than replacing it. In this structure, the SIEM acts as the evidential record of events, while AI analyses data, builds context and supports investigations.

Customer-dedicated local large language models are deployed within sovereign environments and trained on each organisation’s own estate. This is intended to improve accuracy by grounding analysis in local operating conditions while limiting the movement of sensitive security data outside customer-controlled infrastructure.

The launch comes as UK cyber policy places greater emphasis on early threat identification and stronger domestic control over defensive tools. e2e-assure linked the product to GCHQ’s AI Cyber Shield initiative and to broader concerns about dependence on foreign technology providers for security operations.

“Cumulo represents a shift away from traditional SOC and SIEM environments that are largely human-centric and reactive because they rely on sequential alert triage and retrospective investigation. Instead, Cumulo uses an AI-first security operating system,” said Rob Demain, chief executive officer of e2e-assure.

“Threats are now moving faster than human-led workflows can keep pace with, leaving security teams struggling. At the same time, many AI approaches in security are still constrained by legacy architectures that force them to rebuild context after the fact. We built Cumulo to change that by continuously building understanding as data is generated, while keeping expert analysts at the centre of decision-making,” Demain said.

The service retains a human review model, with SC-cleared security staff involved in decisions rather than allowing the platform to operate autonomously. Customer security and operations teams also remain involved throughout investigations, particularly where risk appetite and operational constraints differ between organisations.

Behind that model is a layered AI structure that separates environment-specific reasoning from broader research and intelligence tasks. A local model layer handles detection and analysis tied to the customer estate, while a separate intelligence layer correlates wider threat data. A further model layer is used for non-sensitive enrichment work.

The platform also uses several AI models to review investigations from different perspectives, creating what e2e-assure calls an auditable view of each alert through its Cumulo Analyst Helper. Findings are then checked against threat intelligence and deterministic detection engines before reaching an analyst, in an effort to reduce false or misleading outputs.

Product tiers

Cumulo is being offered through a tiered model aimed at different levels of security maturity. The standard version includes AI-led investigation, autonomous threat hunting, threat intelligence, centralised reporting and compliance dashboards.

The higher tier adds unified monitoring across IT and OT systems, digital twin functions, live compliance dashboards and cross-environment correlation for organisations with more complex operational estates. e2e-assure says the model is intended to help users identify and rank vulnerabilities across interconnected environments before they are exploited.

The company has provided managed security operations services to government and critical infrastructure customers for more than a decade. Its security operations centre is staffed by UK-based cleared cyber professionals, and the Cumulo platform is fully owned by the business rather than tied to a single third-party technology stack.

“For organisations responsible for critical national infrastructure and essential services such as energy, water, transport, telecommunications and government operations, resilience isn’t just about identifying threats faster; it’s about ensuring your ability to defend remains intact during a crisis,” Demain said.

“As more security capabilities move into the cloud, questions around sovereignty, dependency and operational continuity continue to mount. For organisations operating in regulated or high-dependence environments, reliance on external AI infrastructure can introduce risks around data residency, transparency and continued access to critical defensive capabilities. Cumulo addresses these challenges by keeping sensitive operational knowledge within customer-controlled environments, reducing exposure to external disruption and helping organisations maintain visibility and cyber defence capability even during major incidents, connectivity outages or wider infrastructure disruption,” he said.

QuEra Computing has announced Libra, its first fault-tolerant quantum computer, and plans to make the system available on Amazon Braket in 2028.

It has also expanded its multi-year strategic collaboration with Amazon Web Services, with Libra becoming the first system covered by the broader agreement.

The Boston-based quantum computing company describes Libra as a megaquop-class machine, a term it uses for a system intended to carry out about one million reliable logical quantum operations. The machine is projected to offer more than 256 error-corrected logical qubits and a logical error rate of 10−6.

Fault-tolerant quantum computing is widely viewed as a key step beyond today’s error-prone systems because it is intended to support longer, more dependable calculations. QuEra says this could open a path to early commercial and research workflows in fields such as molecular simulation, materials discovery and optimisation, where classical computing methods can struggle as problems grow larger.

AWS link

Under the arrangement, AWS customers will be able to access Libra through Amazon Braket, the cloud group’s managed quantum computing service. AWS says Braket gives users a single environment to build and run quantum applications alongside existing classical infrastructure, including high-performance computing, artificial intelligence and machine learning services.

The tie-up extends a relationship that began in 2022, when QuEra’s Aquila system became available on Amazon Braket. Aquila is a 256-physical-qubit neutral-atom quantum computer, while QuEra’s Gemini system, which the company says has logical-qubit functions, is co-located with the ABCI-Q supercomputer in Japan.

Andy Ory, chief executive officer of QuEra Computing, said the company sees the announcement as part of a broader shift in the industry.

“Fault-tolerant quantum computing is moving from a scientific milestone to an engineering and deployment roadmap,” Ory said.

“We have executed this roadmap in the open, with peer-reviewed milestones and validated system advances. Libra brings fault-tolerant computing to the cloud at scale in 2028. It is an important step forward, and subsequent generations will scale even further, as we will reveal in our roadmap webinar later this month. We are inviting leaders to engage now so they can build the talent, use cases and workflows needed to be ready when these systems come online,” he said.

Research base

QuEra says every building block of the Libra architecture has already been validated in peer-reviewed research. It points to eight papers in Nature and Physical Review Letters by its teams and by researchers in the laboratories of its scientific founders at Harvard and MIT.

According to the company, the papers cover logical qubits, below-threshold error correction, transversal logical operations, fast decoding for real-time error correction, sustained operation of thousands of qubits with continuous atom reloading, and error-correcting codes intended to reduce the number of physical qubits needed for each logical qubit.

Neutral-atom quantum computing has drawn increasing attention as one of several technical approaches in the race to build practical quantum systems. QuEra’s strategy has focused on demonstrating error correction and then scaling towards fault tolerance, rather than only increasing raw qubit counts.

Amazon Web Services says the collaboration reflects its view that fault-tolerant quantum systems will become part of customers’ computing environments.

“We believe fault-tolerant quantum computing will become a foundational part of how customers solve their hardest computational problems on AWS. QuEra’s technology has demonstrated a clear path to that future. By bringing these capabilities to customers through Amazon Braket, they can combine QuEra’s fault-tolerant quantum processors with the scalable AWS HPC and AI services they already rely on,” said Eric Kessler, general manager of Amazon Braket at AWS.

Commercial pressure

QuEra is also using the announcement to urge potential users to prepare for fault-tolerant systems before they become commercially available. It plans to keep building successive in-house generations of fault-tolerant systems ahead of Libra’s release, both to refine the design and to give selected partners earlier access to working environments.

Yuval Boger, chief commercial officer of QuEra, said organisations that delay planning could lose time once the systems arrive.

“Waiting until 2028 to build a quantum strategy is a competitive risk,” Boger said.

“The algorithms that will harness fault-tolerant systems at this scale might not yet exist. Given that Libra will be available on the cloud in 2028 with a one-in-a-million error rate, the organizations that start co-developing now will be operational on day one, not catching up,” he said.

Industry analysts say the announcement marks an important moment for a field often criticised for setting ambitious targets without enough technical disclosure. QuEra’s emphasis on peer-reviewed milestones appears intended to distinguish its timetable from less transparent claims in the sector.

“QuEra’s plan to deliver fault-tolerant systems in 2028 represents a significant inflection point for the quantum computing industry. QuEra’s approach entails publishing every milestone, validating through peer review and now offering concrete QC end-user engagement paths. This disciplined and visible strategy is what aspiring QC end users in HPC centres and related government programs want to see before committing substantial resources to an emerging technology,” said Bob Sorensen, chief analyst for quantum computing at Hyperion Research.