UK BUSINESS AI UNIVERSE: Labs: Corporate research labs

Owner or parent

Amazon

Site and address as published

"Waverly Gate" (page spelling), Edinburgh city centre, in the shell of the former General Post Office building, eight floors

What it does

Amazon scientists in Edinburgh research a range of Amazon products and services. They work in mixed teams with engineers, product managers and designers, and the page says they own solutions from research to production. Listed areas: AI interpretability, explainable AI, computer vision, advertising, recommender systems, deep learning and reinforcement learning.

Research groups or focus

AI interpretability; computer vision; explainable AI; advertising; recommender systems; deep learning; reinforcement learning

Status now

Active, per the location page read on 31 Oct 2026.

Labs Round 8 deep profile (31 Oct 2026), as the researchers wrote it; confidence: one source.

Sources

1. amazon.science/locations/edinburgh

Owner or parent

Amazon

Founded or opened

London, Principal Place, opened 2017; Cambridge not stated

Other leaders

Emine Yilmaz — UCL professor and Amazon Scholar; leads a London-based Alexa voice-shopping research team — Oct 2020; Yoelle Maarek — heads Alexa Shopping Research and Science — Oct 2020 (both from a 2020 page; current roles not checked)

Site and address as published

Principal Place, Shoreditch, London (15 floors; "nearly 5,000 seats", which is capacity, not staff); Cambridge office (no address)

What it does

Amazon scientists in London and Cambridge work on machine learning, natural language processing, text-to-speech, computer vision, search, contextual advertising, pricing, data security, supply chain, logistics and automated reasoning. The products named are Amazon retail, Prime Video, Alexa, Ring, Halo and Echo Show.

Research groups or focus

data science; machine learning; natural language processing; text-to-speech; marketing; content pricing; data security; search; contextual advertising; computer vision; supply chain optimisation; transportation logistics; automated reasoning

Notable outputs

Not stated (none named on the location page)

Status now

Active, per the location page read on 31 Oct 2026; no dated 2025–26 London or Cambridge news was opened.

Labs Round 8 deep profile (31 Oct 2026), as the researchers wrote it; confidence: one source.

Sources

1. amazon.science/locations/london-and-cambridge
2. amazon.science/working-at-amazon/emine-yilmaz-an

Owner or parent

Thomson Reuters and Imperial College London

Founded or opened

announced 2 Dec 2025 (Toronto dateline); five-year partnership

Lab director or lead

Alessandra Russo — Lab Director (per photo notes); Professor in Applied Computational Logic, Imperial Department of Computing; Co-director of the School of Convergence Science, Human and Artificial Intelligence — from Dec 2025

Other leaders

Jonathan Richard Schwarz — Lab Associate Director (Head of AI Research, Thomson Reuters; Visiting Professor at Imperial); Felix Steffek — Lab Associate Director (Professor of Law, University of Cambridge); Mary Ryan — Vice Provost for Research and Enterprise, Imperial

Site and address as published

Imperial College London, "a dedicated facility in Imperial's deep tech community"

Headcount

"more than a dozen PhD students" — 2 Dec 2025 — thomsonreuters.com (single source)

What it does

A five-year research lab at Imperial College London funded by Thomson Reuters. More than a dozen PhD students work with Thomson Reuters scientists and Imperial academics, using Imperial's computing cluster. Topics: AI agents, data-centred machine learning, retrieval and grounding, reasoning and planning, evaluation and safety, and human-in-the-loop work, plus AI's effects on the economy, work and access to justice.

Research groups or focus

agentic systems; data-centric machine learning; retrieval and grounding; reasoning and planning; evaluation and safety; human-in-the-loop workflows; economic impact; future of work; access to justice

Funding or budget as published

"guaranteed multi-year funding", amount not disclosed — 2 Dec 2025 (already DEAL-LAB-6001)

Partners named

NVIDIA — its $250m UK compute commitment names Imperial College London, quoting Russo as "director of Imperial Frontier AI Research Lab" — 13 Oct 2026

Status now

Active: its director was quoted as a named partner in NVIDIA's UK compute commitment on 13 Oct 2026.

Labs Round 8 deep profile (31 Oct 2026), as the researchers wrote it; confidence: verified (Thomson Reuters release and NVIDIA release; UKTN already on record).

Sources

1. thomsonreuters.com/en/press-releases/2025/decemb
2. nvidianews.nvidia.com/news/nvidia-commits-250-mi

Owner or parent

Graphcore, a wholly owned subsidiary of SoftBank Group Corp (acquired 11 Jul 2024)

Founded or opened

the lab was announced on 20 Nov 2025. Graphcore was co-founded in 2016 by Nigel Toon and Simon Knowles.

Lab director or lead

Marcus McElroy — General Manager, "at the helm" — from 31 Jul 2026

Other leaders

executive team Nick Bishop; Helen Byrne; Sally Doherty; Tim Ramsdale; Rami Sinno; John Walsh (titles not given) — Jul 2026; Nigel Toon — co-founder, stepped down as Executive Chair on 31 Jul 2026

Site and address as published

headquarters in Bristol, moving into "a purpose built global headquarters in Bristol, UK" in Sep 2026; offices in Cambridge, London, Gdansk and Hsinchu (Jul 2024)

Headcount

lab "expected to double headcount to 750 jobs" — 20 Nov 2025 — GOV.UK (single source); company "approaching 1,000 employees" — 31 Jul 2026 — graphcore.ai; 75 new jobs, a 20% rise — 7 Nov 2024 — graphcore.ai

What it does

Graphcore is a Bristol company that builds silicon, hardware and software for AI compute, owned by SoftBank since 2024. In November 2025 the UK government said Graphcore and SoftBank would open a new AI development lab in Bristol, doubling headcount there to 750. In July 2026 Graphcore said it was approaching 1,000 staff and moving to a new Bristol headquarters.

Research groups or focus

"silicon design, hardware, datacenter infrastructure, software, and AI research" (hiring areas, Nov 2024)

Notable outputs

no new outputs found; earlier chips are already on the site (GC1, GC200, Bow IPU, BowPod 1024, IPU-M2000)

Funding or budget as published

Not stated (no acquisition price or lab budget published on pages opened)

Partners named

SoftBank Group — owner — Jul 2024; SoftBank Investment Advisers (Vikas J. Parekh, Managing Partner) — quoted — Jul 2024

Uk government links

GOV.UK announcement of 20 Nov 2025 ("Graphcore and SoftBank renew their backing for the UK")

Status now

Active: Marcus McElroy has led the company since Nigel Toon stepped down on 31 Jul 2026, and Graphcore said it would move into a new Bristol headquarters in Sep 2026.

Labs Round 8 deep profile (31 Oct 2026), as the researchers wrote it; confidence: one source (lab and 750 jobs: GOV.UK only); company facts verified across Graphcore pages.

Sources

1. gov.uk/government/news/ai-to-power-national-rene
2. graphcore.ai/posts/graphcore-co-founder-and-exec
3. graphcore.ai/posts/graphcore-joins-softbank-grou
4. graphcore.ai/about
5. graphcore.ai/posts/graphcore-begins-hiring-drive

Owner or parent

IBM Research (IBM Research Europe)

Lab director or lead

Michael Bradley — UK Senior Research Manager, IBM Research (Hursley) — date not stated

Other leaders

Helen Stanton — Program Manager for Research; Dave Braines — Delivery lead for Safer Materials Advisor; Geeth de Mel — Manager and Senior Research Scientist; Anna Paola Carrieri — Manager and Senior Research Scientist

Site and address as published

"Based at Hursley"; Daresbury Lab (joint Hartree National Centre for Digital Innovation with STFC)

Headcount

Not stated ("growing team")

What it does

IBM's UK research lab, based at Hursley, with a team at Daresbury working with STFC's Hartree Centre. Its teams build algorithms for AI and quantum applications in earth and space observation, healthcare and life sciences, materials and AI-accelerated science. In 2026 it released TokaMind, a foundation model of fusion plasma built with the UK Atomic Energy Authority.

Research groups or focus

"artificial intelligence, climate and life sciences, quantum computing, and hybrid cloud"; Bradley's teams: Earth and Space Observation, Healthcare and Life Sciences, AI-accelerated science; projects: physics extremes computing, simulation for accelerated discovery, Auto-omics

Notable outputs

TokaMind — fusion plasma foundation model (about 9 million parameters, about 40 signals from the MAST tokamak), with UKAEA and the Hartree Centre — 23 Apr 2026 (partnership already BIZ-CLD-6028); TokaMark — benchmark of 14 plasma-analysis tasks — 2026; flood-event detection from satellite data, co-developed with Hartree — blog of 31 Mar 2023; AI platform for inflammatory bowel disease drug development with Reprocell and Hartree — 2023. The UK lab page also lists TerraStackAI, an IBM–ESA open-source flood and wildfire model, an open-source ocean model and foundation models for materials; UK authorship of each is not stated.

Selected papers

TokaMind (full title not seen) — arXiv — Feb 2026 — https://arxiv.org/html/2602.15084v1 ; TokaMark (full title not seen) — arXiv — Feb 2026 — https://arxiv.org/html/2602.10132v2 (both links taken from IBM's blog; arXiv pages not opened)

Funding or budget as published

Hartree National Centre for Digital Innovation, £210m over five years — 4 Jun 2021 — https://uk.newsroom.ibm.com/2021-06-04-UK-STFC-Hartree-Centre-and-IBM-Begin-Five-Year-210-Million-Partnership-to-Accelerate-Discovery-and-Innovation-with-AI-and-Quantum-Computing (already DEAL-BIG-001)

Partners named

STFC Hartree Centre — HNCDI, "our first Discovery Accelerator in Europe" — Jun 2021; UKAEA — TokaMind — Apr 2026; Reprocell — IBD drug-development platform — 2023; Unilever and Gexcon UK — HNCDI launch partners quoted — Jun 2021

Uk government links

HNCDI with STFC (UKRI), announced with Science Minister Amanda Solloway, Jun 2021; DSIT's Regulatory Innovation Office co-hosted an AI hackathon with IBM (sector plan update, 10 Jun 2026)

Status now

Active; HNCDI's five-year term began June 2021, and no renewal was found on pages opened.

Labs Round 8 deep profile (31 Oct 2026), as the researchers wrote it; confidence: verified (HNCDI: IBM newsroom and IBM Research blog); other fields one source.

Sources

1. research.ibm.com/labs/uk
2. research.ibm.com/people/michael-bradley
3. uk.newsroom.ibm.com/2021-06-04-UK-STFC-Hartree-C
4. research.ibm.com/blog/ibm-stfc-accelerating-inno
5. research.ibm.com/blog/fusion-plasma-ai-model

Owner or parent

Meta (Fundamental AI Research, FAIR)

Lab director or lead

Not stated for London. Naila Murray — head of FAIR EMEA (regional role) — Jun 2023

Other leaders (historical; current roles not checked)

Sebastian Riedel — Facebook AI Research Manager, dual UCL affiliation — Feb 2021; Tim Rocktäschel and Edward Grefenstette — FAIR London researchers with dual UCL affiliation — Feb 2021; Andrea Vedaldi — joined the London office (also Oxford associate professor) — Jul 2018

Site and address as published

London (no address)

What it does

Meta's Fundamental AI Research team in London, one of four EMEA hubs with Paris, Tel Aviv and Zurich. Pages opened show London work on computer vision and 3D (Implicitron, open-sourced with Paris in 2022), on language (the Bloomsbury AI team joined in 2018), and a UK PhD programme with UCL from 2021. No 2025–26 page about the London team was found.

Research groups or focus (shared by the four EMEA hubs)

self-supervised learning; reinforcement learning; speech and audio; computer vision; natural language modelling; responsible AI; machine learning theory; model efficiency; AR/VR

Notable outputs

Implicitron — modular framework in PyTorch3D that learns 3D objects or scenes from sparse images, open-sourced by London and Paris researchers — Aug 2022

Partners named

University College London — FAIR UK PhD programme, four students in year one — Feb 2021; University of Oxford — through Andrea Vedaldi — Jul 2018

Uk government links

DSIT's sector plan update (10 Jun 2026) names Meta among AI firms in London's King's Cross cluster

Status now

Not stated: no 2025–26 page about the London research hub was opened.

Labs Round 8 deep profile (31 Oct 2026), as the researchers wrote it; confidence: one source.

Sources

1. about.fb.com/news/2023/06/a-spotlight-on-the-fou
2. ai.meta.com/blog/facebook-ai-research-expands-wi
3. ai.meta.com/blog/fair-unveils-uk-phd-program-in-

Owner or parent

Microsoft, Microsoft AI division (Mustafa Suleyman, EVP and CEO of Microsoft AI)

Founded or opened

announced 7 Apr 2024

Lab director or lead

Jordan Hoffmann — hub lead (formerly Inflection and DeepMind) — from the 7 Apr 2024 announcement (not re-confirmed in 2026)

Other leaders

Mustafa Suleyman — EVP and CEO, Microsoft AI; leads the MAI Superintelligence Team (Nov 2025 post)

Site and address as published

"central London", with a team at Microsoft's Paddington office (Apr 2024)

Headcount

Not stated for the hub. Microsoft UK as a whole: "6,000 employees" — 16 Sep 2025 — https://blogs.microsoft.com/on-the-issues/2025/09/16/microsoft-30-billion-uk-ai-future/. On 31 Oct 2026 Microsoft AI listed 21 London roles (15 Microsoft AI, 6 Health) — https://microsoft.ai/job-region/london-united-kingdom/

What it does

Microsoft AI's research and engineering hub in London. When it was announced in 2024, its stated job was work on state-of-the-art language models, their supporting infrastructure and tooling for foundation models. Its London job listings in October 2026 cover pretraining data, safety post-training, multimodal safety, frontier risk and alignment, AI supercomputing and health.

Research groups or focus

"advance state-of-the-art language models and their supporting infrastructure" (2024). London roles on 31 Oct 2026: Pretraining Data; Safety Post-training; Multimodal Safety; Production Safety; Research Scientist, Frontier Risk & Alignment, Futures team; Compute Infra HPC; AI Supercomputing; Health (Applied AI, Privacy, Product)

Notable outputs (Microsoft AI division-wide; no page credits London)

MAI-Thinking-1 — reasoning model, sparse Mixture of Experts, about 35B active and 1T total parameters, public preview on Microsoft Foundry — 12 Aug 2026; seven in-house MAI models (MAI-Thinking-1, MAI-Code-1-Flash, MAI-Image-2.5, MAI-Image-2.5-Flash, MAI Transcribe-1.5, MAI-Voice-2, MAI-Voice-2-Flash) — 2 Jun 2026

Funding or budget as published

Not stated for the hub. Microsoft's $30bn UK commitment for 2025–2028 is already DEAL-BIG-021

Partners named

OpenAI — named as a collaborator of Microsoft AI London — Apr 2024; Surge — evaluation partner for MAI-Thinking-1 (division-wide) — Aug 2026

Uk government links

HM Treasury release of 10 May 2024 cites the hub ("Microsoft recently announced a new AI hub in London"); AISI–Microsoft partnership of 5 May 2026 (Microsoft-wide; record BIZ-SAFE-2086)

Status now

Active: Microsoft AI listed 21 open London roles on 31 Oct 2026.

Labs Round 8 deep profile (31 Oct 2026), as the researchers wrote it; confidence: verified (opening: Microsoft and GOV.UK); lead one source.

Sources

1. blogs.microsoft.com/blog/2024/04/07/announcing-n
2. gov.uk/government/news/leading-tech-firms-invest
3. microsoft.ai/job-region/london-united-kingdom/
4. microsoft.ai/news/introducing-mai-thinking-1/
5. microsoft.ai/news/building-a-hillclimbing-machin

Owner or parent

Microsoft Research

Founded or opened

Not stated (not stated on the lab page)

Lab director or lead

Aditya Nori — Vice President and Lab Director — date not stated

Other leaders

Austin Donnelly — Senior Principal Research Engineer Lead; Hitesh Ballani — Partner Research Manager; Katja Hofmann — Partner Research Manager; Siân Lindley — Senior Principal Research Manager; Ted Meeds — Senior Principal Research Manager; Rachel Howard — Director Business Management & Ops

Site and address as published

21 Station Road, Cambridge, CB1 2FB; and 198 Cambridge Science Park, Milton Road, Cambridge CB4 0AB

What it does

Microsoft's research lab in Cambridge. It designs hardware for Microsoft's cloud and AI data centres, including optical computing, storage in glass and datacentre networking and robotics. It also works on machine learning for science, industry and society, and on people-centred AI such as generative models of gameplay, tools that show where digital content comes from, and AI that helps people think.

Research groups or focus

Future AI Infrastructure; Machine Intelligence; People-Centric AI; Research Engineering

Notable outputs

WHAM (World and Human Action Models) — generative models of game worlds and player actions from the Muse programme, built with Xbox Game Studios' Ninja Theory; WHAM-RT is playable in Copilot Labs — 2025; the team moved into the Microsoft.AI product group (editor's note, 23 Sep 2026); AOC — analog optical computer for AI and optimisation workloads — date not stated; Project Silica — cloud storage written into quartz glass with femtosecond lasers — date not stated; Project Provenance — tools showing where digital content comes from — date not stated; Tools for Thought — date not stated; Teachable AI Experiences (Tai X) — date not stated; Project Volano — date not stated; CRAFT (datacentre robotics) — date not stated; Project Egypt (greenhouse-gas data) — date not stated; Memora — memory representation for AI agents — 29 Jun 2026; MicroLED datacentre networking — 17 Mar 2026

Selected papers

"WHAM: World and Human Action Models" (title as listed by Microsoft) — Nature — 2025 — https://www.nature.com/articles/s41586-025-08600-3

Partners named

Xbox Game Studios' Ninja Theory — WHAM — 2025; Ottobock — prosthesis image dataset for AI — Jul 2026

Uk government links

Not stated on pages opened

Status now

Active: the lab's news page lists items up to 30 Jul 2026.

Labs Round 8 deep profile (31 Oct 2026), as the researchers wrote it; confidence: one source (the lab's own pages).

Sources

1. microsoft.com/en-us/research/lab/microsoft-resea
2. microsoft.com/en-us/research/people/adityan/
3. microsoft.com/en-us/research/lab/microsoft-resea
4. microsoft.com/en-us/research/project/wham/
5. microsoft.com/en-us/research/lab/microsoft-resea
6. blogs.microsoft.com/on-the-issues/2025/09/16/mic

Owner or parent

Nokia (Nokia Bell Labs)

Lab director or lead

Bo Olofsson — "Software Innovations for Industry Verticals & Site Leader, Nokia Bell Labs Cambridge" — date not stated

Other leaders

Fahim Kawsar — Founding Director of Pervasive Systems research, Cambridge, UK; Michael Eggleston — Data & Devices Group Leader (lab listed at New Providence, New Jersey and Cambridge)

Site and address as published

Broers Building, 21 J.J. Thomson Avenue, Cambridge, CB3 0FA (West Cambridge campus)

What it does

Nokia Bell Labs' site on the University of Cambridge's West Cambridge campus. It works on multimodal AI and machine learning for connected devices and wearables, on social computing and on materials science. It is one of two sites of Bell Labs' Data and Devices lab, which builds sensors and AI for health and industrial monitoring.

Research groups or focus

"multimodal artificial intelligence/ machine learning" for IoT and wearables; social computing; materials science; Pervasive Systems research; Data and Devices Lab (AI Research Lab); founding member of the Research Centre for Mobile, Wearable Systems and Augmented Intelligence

Notable outputs (Data and Devices lab, two sites; the site of each project is not stated)

The Patchkeeper — wearable monitor with ECG, PPG, IMU and electronic stethoscope; The Gecko — compact solar-powered environment sensor; OCT-based biochemical sensing; battery-free backscatter RFID asset tracking — dates not stated

Selected papers (Data and Devices lab page, which covers both sites)

"Quantified Canine: Inferring Dog Personality From Wearables" — CHI '23 — 2023 — https://dl.acm.org/doi/10.1145/3544548.3581088 ; "Heart Rate Extraction from Abdominal Audio Signals" — ICASSP 2023 — 2023 — https://ieeexplore.ieee.org/abstract/document/10096600 ; "Stress Inference from Abdominal Sounds using Machine Learning" — EMBC 2022 — 2022 — https://ieeexplore.ieee.org/abstract/document/9871165 ; "The Gecko Sensor: An Ultra-Compact, Low-Cost, Solar-Powered Environment Monitoring Device" — IEEE Sensors 2022 — 2022 — https://www.nokia.com/sites/default/files/2024-11/the_gecko_sensor_an_ultra-compact_low-cost_solar-powered_environment_monitoring_device.pdf ; "ComFeel" — Proc. ACM IMWUT vol. 4 no. 4 — 2020 — https://www.nokia.com/sites/default/files/2024-11/comfeel.pdf

Partners named

University of Cambridge — the research centre above; University of Oxford; UCL; Lancaster; Newcastle; King's College London — academic partners as listed on the site page

Status now

Active, per the Nokia location page read on 31 Oct 2026.

Labs Round 8 deep profile (31 Oct 2026), as the researchers wrote it; confidence: one source.

Sources

1. nokia.com/bell-labs/about/locations/cambridge-uk
2. nokia.com/people/fahim-kawsar/
3. nokia.com/bell-labs/research/air-lab/data-and-de

Owner or parent

NVIDIA

Founded or opened

Not stated (an existing lab by June 2025)

Site and address as published

Bristol (no address)

What it does

An NVIDIA AI lab in Bristol. In June 2025 NVIDIA told the UK government it would expand the lab to other parts of the UK to speed up UK AI research. None of the NVIDIA pages I opened describe the lab's own work. NVIDIA's UK research links include the Isambard-AI supercomputer at the University of Bristol, the UK-LLM project with UCL and Bangor University, and compute for UK agentic-safety labs.

Research groups or focus

"to accelerate UK research in AI" (GOV.UK, Jun 2025)

Notable outputs (UK-wide, not credited to the Bristol lab)

UK-LLM — Welsh and English model with UCL and Bangor University, built on NVIDIA Nemotron — Sep 2025; electrostatics-aware foundation models with University of Cambridge researchers, using cuEquivariance — Sep 2025 (Nemotron-Personas-UK-Extended is already BIZ-LAB-7098)

Funding or budget as published

Not stated for the lab. NVIDIA committed $250m in compute donations over five years ($50m a year) to UK agentic safety and security research — 13 Oct 2026 — https://nvidianews.nvidia.com/news/nvidia-commits-250-million-to-advance-uk-agentic-safety-and-security-research-over-the-next-five-years

Partners named

BOLD Lab (Oxford), SOFAIR Lab (UCL), Imperial College London — compute donation — Oct 2026; University of Bristol — Isambard-AI on Grace Hopper — Jul 2025; Sovereign AI Industry Forum founding members Babcock, BAE Systems, Barclays, BT, National Grid, Standard Chartered — Jun 2025

Uk government links

two MoUs with the UK government (Jun 2025): an AI talent pipeline, and university research on AI in advanced connectivity; DSIT update (10 Jun 2026): JOINER terminals in Leeds, Bristol and Edinburgh awarded NVIDIA GPU platforms

Status now

No page opened gives a 2026 update on the Bristol lab itself; NVIDIA's wider UK research activity continued with the October 2026 compute commitment.

Labs Round 8 deep profile (31 Oct 2026), as the researchers wrote it; confidence: one source (the lab itself is named only by GOV.UK).

Sources

1. gov.uk/government/news/pm-launches-national-skil
2. nvidianews.nvidia.com/news/nvidia-commits-250-mi
3. blogs.nvidia.com/blog/uk-partner-ecosystem-ai-ma
4. blogs.nvidia.com/blog/uk-ai-vision/

Owner or parent

OpenAI

Founded or opened

first international office, London, announced 28 Jun 2023; named a major research hub on 26 Feb 2026

Lab director or lead

Phoebe Thacker — global head of data research programmes and London site lead — Apr 2026

Other leaders

Mark Chen — chief research officer, OpenAI (quoted Feb 2026); Diane Yoon — VP of People (2023)

Site and address as published

new office at Regent Quarter, King's Cross (Jahn Court and the Brassworks Building), 88,500 sq ft, room for up to 544 staff, opening 2027

Headcount

Not stated (544 is capacity, not staff)

What it does

OpenAI's London base, its first office outside the US (2023). In February 2026 OpenAI said London would become its biggest research hub outside San Francisco, with London researchers owning "key components" of frontier model development, including quality, reliability, alignment and safety. Its Training team hires London researchers to build OpenAI's main language models.

Research groups or focus

Training team ("responsible for producing the large language models that power our research"); frontier model development: quality, reliability, alignment and safety

Notable outputs

Not stated (none credited to London)

Partners named

Endurance Land — developer of the King's Cross office — Apr 2026

Uk government links

OpenAI–UK MoU signed by Sam Altman and Peter Kyle on 21 Jul 2025 (record BIZ-BIG-003); Technology Secretary Liz Kendall welcomed the research hub (Feb 2026); £5.6m to AISI's Alignment Project (Feb 2026); £100m in compute and model credits for the AI Biosecurity Lab (Oct 2026)

Status now

Active and growing; its permanent King's Cross office is due in 2027.

Labs Round 8 deep profile (31 Oct 2026), as the researchers wrote it; confidence: verified (OpenAI pages and UKTN).

Sources

1. uktech.news/ai/openai-names-london-as-its-next-m
2. uktech.news/ai/openai-secures-first-permanent-lo
3. openai.com/careers/researcher-training-london-lo
4. openai.com/index/introducing-openai-london/
5. uktech.news/ai/openai-expands-uk-footprint-and-b

Owner or parent

Thomson Reuters (acquired Aug 2024; terms undisclosed)

Founded or opened

2022

Lab director or lead

Not stated (now part of Thomson Reuters)

Other leaders

Alexander Kardos-Nyheim — co-founder; Dr Jonathan Schwarz — co-founder (now Head of AI Research, Thomson Reuters)

Headcount

"15-person startup" when acquired — per Thomson Reuters, Dec 2025 (single source)

What it does

A London start-up that trained large language models for legal work, with a team "drawn from Cambridge, DeepMind, Harvard and MIT". Thomson Reuters bought it in August 2024 to support its CoCounsel assistant. Its team became the core of TR Labs' Foundational Research group, which later built Thomson Reuters' own model, Thomson.

Research groups or focus

legal-domain large language models

Notable outputs

Safe Sign's legal LLMs (unnamed); successor: Thomson model — Jul 2026 (see BIZ-LAB-6004)

Funding or budget as published

Not stated (investors not stated)

Status now

Acquired; the team works as TR Labs Foundational Research inside Thomson Reuters.

Labs Round 8 deep profile (31 Oct 2026), as the researchers wrote it; confidence: verified (acquisition: UKTN and Thomson Reuters); founding year one source (UKTN).

Sources

1. uktech.news/ai/thomson-reuters-acquires-uk-ai-st
2. thomsonreuters.com/en-us/posts/innovation/from-s

Owner or parent

Samsung Research (Samsung Electronics). Samsung's AI Center network (SAIC) was "established in November 2017".

Founded or opened

Not stated for Cambridge (Andrew Blake is named as its founding head)

Lab director or lead

Timothy Hospedales — "Head of Europe AI Center" and "Head of AIC-Cambridge" — current centre page. The page also carries an older line naming Andrew Blake as leader.

Other leaders

Georgios Tzimiropoulos — "Leader of AIC-Cambridge"; Andrew Blake — founding Head (former director of Microsoft Research Cambridge; former Director at the Alan Turing Institute)

Site and address as published

"Located in Cambridge, UK" (no street address; contact cambridge.ai@samsung.com)

What it does

Samsung's AI research centre in Cambridge. It does open research on vision and language, efficient foundation models that run on devices, speech, and machine learning. It moves results into Galaxy phones, for example on-device personalised sound enhancement that helps Bixby hear its owner's voice in noise.

Research groups or focus

Future Interaction (Vision & Language, Large Scale Training, Video Understanding, Diffusion Models); Embedded AI (Efficient Foundation Model Architectures, AutoML, Model Compression, Federated Learning, Speech Recognition & Generation, NPU Hardware Design); Machine Learning & Data Intelligence (Meta-Learning, In-context Learning and LLMs, Bayesian Neural Networks, Neuro-Symbolic Learning, Neural Rendering)

Notable outputs

Personalized Sound Enhancement — on-device, real-time audio cleaning before Bixby speech recognition; shipped on Galaxy S22–S23, Fold3–Fold4 and Flip3–Flip4, then wider devices including Fold5 and Flip5 — date not stated

Selected papers

"Meta-Learning in Neural Networks: A Survey" — IEEE T-PAMI — 2020 — https://research.samsung.com/aicenter_cambridge ; "Learning to Compare: Relation Network for Few-Shot Learning" — CVPR — 2018 — https://research.samsung.com/aicenter_cambridge ; "Conditioning Sequence-to-sequence Networks with Learned Activations" (Ramos, Mehrotra, Lane, Bhattacharya) — ICLR — 2022 — https://research.samsung.com/blog/Personalized-Sound-Enhancement-A-Journey-From-Idea-To-Deployment-Into-The-Hands-Of-Millions-Of-Samsung-Users

Status now

Active, per the centre page read on 31 Oct 2026; no dated 2025–26 news item was opened.

Awards

Andrew Blake — IEEE Longuet-Higgins Prize 2021 (for a 2011 Microsoft Research paper)

Labs Round 8 deep profile (31 Oct 2026), as the researchers wrote it; confidence: one source.

Sources

1. research.samsung.com/aicenter_cambridge
2. research.samsung.com/news/Head-of-Samsung-AI-Cen
3. research.samsung.com/blog/Personalized-Sound-Enh
4. research.samsung.com/aicenter

Owner or parent

Toshiba Europe Limited (part of it since April 2020)

Founded or opened

1991, as Toshiba Cambridge Research Centre; "Toshiba's first overseas corporate-level R&D laboratory"

Lab director or lead

Roberto Cipolla — has led the lab since July 2007 (current title not given)

Other leaders

Sir Michael Pepper — led the lab from 1991 to July 2007; Chao Zhang — multimodal foundation models, Vision & Learning Group; Rudra Poudel — lead scientist, World Models for Reinforcement Learning, Vision & Learning Group

Site and address as published

Cambridge (no address on pages opened)

What it does

Toshiba's research lab in Cambridge, founded in 1991. It works on quantum communications and devices, speech and language, and computer vision and learning. Since 2024 it has expanded "embodied AI": systems that adapt quickly to new places, keep learning, and take instructions from people in plain language. Toshiba is putting £15m into this over five years and plans a first industrial prototype in 2027.

Research groups or focus

Quantum Information Group (Quantum Communications Research; Quantum Devices Research); Embodied AI: Language & Interaction Group and Vision & Learning Group

Notable outputs

Whisma — speech-LLM for zero-shot spoken language understanding — 2024; ReCoRe — training framework for world models — CVPR 2024; dialogue-based localisation (pose estimation through natural language) — CVPR 2024; LUCES and a Toshiba CAD dataset (linked, no details) — dates not stated

Selected papers

"Whisma: A speech-LLM to perform zero-shot spoken language understanding" — IEEE SLT — 2024 — https://www.toshiba.eu/cambridge-research-laboratory/embodied-ai/ ; "Recurrent Reinforcement Learning with Memoroids" — NeurIPS — 2024 — same url ; "NPLMV-PS: Neural Point-Light Multi-View Photometric Stereo" — WACV — 2025 — same url ; "An LLM Feature-based Framework for Dialogue Constructiveness Assessment" — EMNLP — 2024 — same url ; "ToolReAGt: Tool Retrieval for LLM-based Complex Task Solution via RAG" — ACL 2025 KnowFM workshop — 2025 — same url ; "Conditional Multi-Stage Failure Recovery for Embodied Agents" — ACL 2025 REALM workshop — 2025 — same url

Funding or budget as published

£15m from Toshiba over five years for embodied AI — 17 Jun 2024 — https://www.toshiba.eu/cambridge-research-laboratory/news/toshibas-leading-cambridge-research-laboratory-crl-doubles-down-on-cutting-edge-embodied-ai-to-bring-ai-into-the-real-world/ (single source)

Partners named

University of Cambridge — Cavendish Laboratory and Engineering Department — ongoing

Status now

Active; the first industrial embodied-AI prototype is planned for 2027.

Labs Round 8 deep profile (31 Oct 2026), as the researchers wrote it; confidence: one source.

Sources

1. toshiba.eu/cambridge-research-laboratory/about-u
2. toshiba.eu/cambridge-research-laboratory/news/to
3. toshiba.eu/cambridge-research-laboratory/embodie

Owner or parent

Thomson Reuters (TR Labs)

Founded or opened

built from Safe Sign Technologies after Thomson Reuters bought it in Aug 2024; "now called Foundational Research" (Dec 2025)

Lab director or lead

Jonathan Schwarz — Head of Foundational Research (Dec 2025) and Head of AI Research, Thomson Reuters (Dec 2025; Jul 2026); co-founder of Safe Sign; previously seven years as a senior researcher at Google DeepMind

Other leaders

Joel Hron — Chief Technology Officer, Thomson Reuters; Wassim Seifeddine — TR Labs researcher

Site and address as published

London is listed among TR research hubs (London, Toronto, Zug, Minneapolis); the team's own base is not stated on pages opened

Headcount

"50-person global group" (grown from a "15-person startup") — 12 Dec 2025 — https://www.thomsonreuters.com/en-us/posts/innovation/from-startup-vision-to-ai-breakthrough-one-year-with-safe-sign/ (single source)

What it does

Thomson Reuters' in-house AI research group, built from London legal-AI start-up Safe Sign Technologies, which it bought in 2024. It trains and evaluates language models for legal and professional work and publishes on data mixtures, model efficiency and evaluation. It built "Thomson", Thomson Reuters' own model, which became the default model for Tabular Analysis in CoCounsel Legal from August 2026.

Research groups or focus

"data mixture optimization, model efficiency, and responsible evaluation methods"

Notable outputs

Thomson (Thomson-1-Large in the benchmark table) — Thomson Reuters' own model for professional and legal work, built on an open-source foundation model and trained on Westlaw, Practical Law, Checkpoint and Reuters content; default model for Tabular Analysis in CoCounsel Legal from Aug 2026 — announced 31 Jul 2026. The company reports it competitive with Claude Opus 4.8 and ahead of GPT-5.5 and Gemini 3.1 Pro on its own early tests.

Selected papers

two NeurIPS 2025 publications (titles not given on the page): data weighting that improves training efficiency "by as much as 500 percent", and evaluation methods that cut benchmarking cost "up to 18 times" — NeurIPS — Dec 2025 — https://www.thomsonreuters.com/en-us/posts/innovation/from-neurips-2025-how-thomson-reuters-is-advancing-the-frontier-of-ai-research/

Partners named

Imperial College London — Frontier AI Research Lab — 2 Dec 2025

Status now

Active: Thomson Reuters launched its own model, Thomson, in summer 2026.

Labs Round 8 deep profile (31 Oct 2026), as the researchers wrote it; confidence: verified (Schwarz's role appears on three Thomson Reuters pages and at UKTN); headcount and model claims one source.

Sources

1. thomsonreuters.com/en-us/posts/innovation/from-n
2. thomsonreuters.com/en-us/posts/innovation/from-s
3. thomsonreuters.com/en-us/posts/innovation/thomso

Owner or parent

xAI

Founded or opened

UK entity incorporated in December (year implied 2024), per UKTN citing Companies House; hiring reported 8 Jan 2025

Lab director or lead

Toby Pohlen — xAI founding member and former DeepMind researcher, "thought to be leading the London team" (UKTN, Jan 2025) — lead, not verified

Other leaders

Jared Birchall — cited as Chief Financial Officer (UKTN, Jan 2025)

Site and address as published

X's office in Piccadilly, London (Jan 2025)

Headcount

Not stated. 7 London roles listed on x.ai on 31 Oct 2026 — https://x.ai/careers/open-roles

What it does

xAI's London engineering team, set up in late 2024 in X's Piccadilly office. xAI described it as engineers working on large-scale distributed systems across data, research and product. On 31 October 2026 xAI listed seven London roles, including a Member of Technical Staff in reasoning on its Model team and product roles for X Money.

Research groups or focus

"large-scale distributed systems across data, research, and product" (Jan 2025); London roles on 31 Oct 2026: Backend Engineer – API (Engineering); Backend Engineer (Infrastructure); Legal Counsel; Member of Technical Staff – Reasoning (Model); Member of Technical Staff – X Money; Mobile Android Engineer; Mobile iOS Engineer (Product)

Partners named

Lewis Silkin — law firm that advised on incorporation — Jan 2025

Status now

Hiring: seven London roles were listed on 31 Oct 2026.

Labs Round 8 deep profile (31 Oct 2026), as the researchers wrote it; confidence: one source.

Sources

1. uktech.news/ai/elon-musks-xai-starts-hiring-in-l
2. x.ai/careers/open-roles