The Alan Turing Institute
@turinginst
Followers
57K
Following
7K
Media
3K
Statuses
11K
We are the Alan Turing Institute, the UK’s national institute for data science and artificial intelligence.
London, England
Joined September 2015
Researchers from the Turing, @AnthropicAI & @AISecurityInst have conducted the largest study of data poisoning to date Results show that as little as 250 malicious documents can be used to “poison” a language model, even as model size & training data grow https://t.co/UPqJKGcLmd
turing.ac.uk
A Turing collaboration with the AI Security Institute and Anthropic will help to protect AI models from misuse
1
4
15
❓Are you a researcher looking to make real-world impact as part of an expert multi-disciplinary team? Apply to join the Turing’s January #DataStudyGroup: Modelling AI competency needs in the UK labour market 📅26 Jan – 6 Feb 2026 ➡️ Apply now: https://t.co/SSaD3YJBsG
0
0
1
🔐 Securing Critical National Infrastructure in a Data-Driven World Read our new report, in partnership with @lloydsregister, introducing a roadmap to use data for more secure, resilient & sustainable infrastructure:
turing.ac.uk
This report examines the role of data and information in managing and maintaining critical infrastructure in the UK.
0
1
3
As model capabilities increase, more work to defend against data poisoning is essential to ensure their secure and trustworthy deployment. ➡️Read the study:
arxiv.org
Poisoning attacks can compromise the safety of large language models (LLMs) by injecting malicious documents into their training data. Existing work has studied pretraining poisoning assuming...
0
0
1
⚠️ Data poisoning occurs when individuals distribute online content designed to corrupt an AI model’s training data, producing potentially harmful outputs. 🚪 This includes the insertion of backdoors – specific phrases that trigger an otherwise-hidden behaviour.
1
0
4
Defence AI Assurance: Identifying promising practice and a system card template for Defence ❓ How can the MoD’s AI policy be implemented in real-world operations? Read the latest report from Turing Senior Research Associate, Anna Knack and co-authors:
turing.ac.uk
The Alan Turing Institute conducted this study in collaboration with Accenture and in coordination with the Ministry of Defence’s Defence AI and Autonomy Uni
1
0
2
💻 Children are facing unprecedented exposure to harmful content online 💡 In a new CETaS Briefing Paper, Sam Stockwell & Rosamund Powell explore how the 🇬🇧, 🇺🇸, 🇪🇺 & 🇦🇺 are using #AgeAssurance tech to tackle this problem, unpacking limits & lessons. https://t.co/vXC1OcH0S6
cetas.turing.ac.uk
1
0
5
📢 Survey call: Data-Centric Engineering community Do you work at the intersection of AI, data science and engineering? 💡 Share your career path, diversity and role insights in our survey: https://t.co/RglvcqZwlv
#DataCentricEngineering #DCE
turing.ac.uk
Participation informationFindings from the survey may be published on The Alan Turing Institute website (see note on anonymity below) and will feed
0
0
1
How is AI changing our lives? 🎧 The Turing's Adrian Weller joined Brian Cox, Neil Lawrence, Steph Wright & Jeanette Winterson to explore this fundamental question. Listen to the latest ep. of the Crick’s A Question of Science podcast: https://t.co/RDNupj7Egk
0
1
4
Explore the country profiles published so far: 🇨🇦 Canada: https://t.co/KAYfPXzq2W 🇪🇺 European Union: https://t.co/LtdOhxe7Ny 🇮🇳 India: https://t.co/YbANNgnvKu 🇸🇬 Singapore:
turing.ac.uk
Singapore’s evolving approach to AI governance prioritises innovation and the strategic deployment of AI technologies in sectors with high socio-economic imp
0
0
1
This project will explore: ✅ Shifting national priorities on AI ✅ Dynamics of cooperation & competition ✅ Potential areas of international alignment
1
0
1
🌍What does the future hold for international AI governance? As the world balances competition with international cooperation & safety, we’re mapping the evolving landscape of AI regulation & standardisation through a series of country profiles:
turing.ac.uk
2
2
4
🌐Enhancing global AI governance Through our work w/ @FCDOGovUK we’re in Hanoi this week hosting a multi-day workshop on AI governance & ethics Supporting a responsible environment for tech dev is key to protecting those at risk while enabling innovation to thrive @UKinVietnam
0
0
1
👥 Persona agents are widely used in roles such as AI assistants & NPCs, but to what extent do they replicate the richness of human behaviour? A new CETaS Expert Analysis explores how psychometrics can – and can’t – provide insights into these agents:
cetas.turing.ac.uk
0
2
2
🔎 Examining AI assurance from a UK Defence perspective New research published today with @AccentureUK explores the assurance of safe and responsible AI in defence, considering real-world military use cases. ➡️Learn more:
turing.ac.uk
The Alan Turing Institute in coordination with the Ministry of Defence (MoD) and
0
1
5
AI for Decarbonisation: Building Sustainable AI webinar 📅6 Oct, 11:30–12:30 ➡️Reg: https://t.co/5je5Nsc7ZC Hear from experts covering practical pathways to decarbonising the AI supply chain: ✅E-waste & dark data ✅Efficient algorithms & data architectures ✅Next-gen hardware
turing.ac.uk
Who should attend? This event will be of interest to professionals and stakeholders across public sector, AI, net zero, cleantech, energy, tran
0
1
3
#TuringConnections PhD Student Presentations & Networking Day 📅24 Oct 2025, 10:00–17:00 📍UCL ➡️Reg by 15 Oct 2025: https://t.co/nyZ3QVbGET 🌟 #DataScience & #AI PhD students: present research, develop skills & connect with peers at this free event. #PhDNetworking
0
0
2
📢Applications open: #DataStudyGroup The challenge: Modelling AI competency needs in the UK labour market 📅26 Jan – 6 Feb 2026 ➡️ Apply now: https://t.co/SSaD3YJBsG Challenge set by the Alan Turing Institute in collaboration with @innovateuk #BridgeAI (supported by @PwC)
1
2
6
What do UK citizens really think about AI-powered data processing for national security? Join CETaS' UK Public Attitudes to National Security Data Processing webinar to explore insights from their novel research. 📅 2 Oct, 15:00–16:00 📍Online ➡️
cetas.turing.ac.uk
As AI and machine learning transform the way national security agencies process vast amounts of information, they bring both crucial opportunities and seriou
0
2
4
Digital identity without compromises. @Zac_Aztec will share how Aztec’s tech is set to revolutionize digital identity without exposing sensitive data. RSVP to attend IRL: https://t.co/c4V7uMi8f4
69
57
278