Instagram story viewer> @theturinginst> Posts

The Alan Turing Institute

@theturinginst

12.3K

followers

158

following

The UK's national institute for #DataScience and #ArtificialIntelligence (#AI)
Organisers of #AIUK and the #TuringLectures
Based at @britishlibrary

POSTS STORIES REELS TAGGED

Download All

Uncertainty about the future should not delay practical action on frontier AI risks

In recent weeks, warnings about the risks posed by the most advanced AI systems have intensified. There is no expert consensus on whether AI poses a long-term existential risk to humanity, and many of the questions involved remain uncertain.

But that uncertainty does not leave us powerless to act.

In our new report, 'Frontier AI Risks: A practical way forward', we identify five areas where practical action is needed now: cyber security, democratic instability, misalignment, loss of meaningful human control and oversight, and chemical and biological threats.

These risks involve capabilities, behaviours and harms that can be observed, studied and tested.

The report also argues that stronger AI verification and whole-system assurance will be an important part of that response, assessing not just individual models, but the technology, people, processes and operating environments around them.

Read the full report at the link in our bio: @theturinginst

#ArtificialIntelligence #AISafety #AIRisk #AIGovernance #AIAssurance #AIVerification #FrontierAI Frontier AI Risks: A practical way forward Uncertainty about the most extreme future AI scenarios should not delay action. by @theturinginst

17

0

16 days ago

A super El Niño is forming. How can AI weather models provide accurate forecasts during anomalous climate events?

Governments and humanitarian organisations rely on accurate seasonal forecasts to prepare for adverse weather events and to help manage water and food security.

We need forecasting tools that can adapt to significant large-scale climate anomalies, like this incoming super El Niño.

While existing physics-based models are grounded in the physical laws governing the atmosphere and oceans, AI models are trained to predict the weather based on past data. We now face the challenge of what happens when we need seasonal predictions for a climate that wasn’t part of AI models’ training material.

The Turing’s Mission Director for Environmental Forecasting, Scott Hosking explores what El Niño could mean for AI forecasting as we start to enter a world it hasn’t seen before (link in bio @theturinginst).

#AI #ElNino #FastNet #WeatherForecasting Text reads: "Is AI ready for El Niño?" over an aerial shot of green coastal mountains. by @theturinginst

29

9

2 months ago

Can AI help us discover new materials?

Aron Walsh, Professor of Materials Design at Imperial College London and Chief Scientific Officer at CuspAI joined us at AI for Science in March this year. He shared insights into how machine learning is accelerating materials discovery by bridging the predictive power of quantum mechanics with the speed of classical simulations.

In his session, he explains:
➡️ key components of materials AI datasets – automation, models and optimisation

➡️ constraints on these datasets, e.g. element availability, toxicity and sustainable computing

➡️ approaches to representing chemical systems and methods to navigate vast chemical spaces

➡️ the role of generative models and agent-based workflows

Watch now – link in bio: @theturinginst

@imperialcollege
#AIforScience #MaterialsDiscovery #QuantumMechanics by @theturinginst

10

0

2 months ago

“Is fusion still 20 years away? With AI, it just might be...”

Lorenzo Zanisi, Lead Data Scientist at the UK Atomic Energy Authority joined us at AI for Science in March this year. He shared insights into his work using AI to accelerate the path to fusion energy, focusing on the STEP spherical tokamak power plant.

In his session, he explains:
➡️ tokamak fusion
➡️ progress toward breakeven
➡️ the need for high fidelity simulation and digital twins that combine experimental diagnostics with simulations

Watch now – link in bio: @theturinginst

#AIforScience #FusionEnergy #Tokamak by @theturinginst

14

0

2 months ago

How do we use AI to predict our changing climate?

Scott Hosking, Mission Director for Environmental Forecasting at the Turing joined us at AI for Science in March this year. He shared insights into the work of the Turing’s Environmental Forecasting mission, including our partnership with the @metoffice on FastNet, an AI weather forecasting model.

In his session, he explores:
➡️ progress toward longer range climate prediction
➡️ use of open-sourced foundation models and UK supercomputing
➡️ the importance of interdisciplinary teams and project management

Watch now – link in bio: @theturinginst

#AIforScience #ClimateChange #FastNet #OpenSource by @theturinginst

20

0

3 months ago

How can foundation models accelerate our understanding of the universe?

Anna Scaife, Turing AI Fellow and Professor of Radio Astronomy at the @universityofmanchester joined us at AI for Science in March this year. She shared insights into her work with foundation models, focusing particularly on the ZooBot model and the Radio Galaxy Zoo model.

In her session, she explores:
➡️ Operationalisation of models
➡️ Possibilities for discovery
➡️ Current limitations in answering some of our biggest questions

Watch now – link in bio: @theturinginst

#AIforScience #FoundationModels #RadioAstronomy by @theturinginst

12

0

3 months ago

AI for Science: watch online

Earlier this year, we brought together leading researchers and practitioners to explore how AI is transforming scientific discovery and innovation.

You can watch all the talks from AI for Science on our YouTube channel – link in bio: @theturinginst

Watch the sessions to learn more about:
✅ AI for physical systems and simulation-based science
✅ Weather forecasting and climate modelling
✅ Fusion energy and materials science
✅ AI as an autonomous scientific collaborator
✅ Trustworthy AI, including safety, interpretability and reliability

#AIforScience #AIResearch by @theturinginst

12

0

3 months ago

The background is primarily white with a bright green polygon at the bottom right. The title text reads: ""Why apply to join a Data Study Group?" Underneath the title is the text: ""Perspectives from previous participants:" with rightward arrow is next to it. by @theturinginst

13

1

4 months ago

🎉Today marks the anniversary of Alan Turing's birth

Alan Turing, our institute's namesake, famously helped crack the Enigma code through codebreaking cyphers during World War II.

To honour his work, on his birthday we have put together the following cypher which we challenge you to decrypt:

🔐 “EBK B JBEADKC CKFMX QRPBVLCPPDCQ BKT EPCBJ?" IMMGDKY RM RAC HSRSPC 76 XCBPQ BYM, RSPDKY IDQRCT B KSJLCP MH RADKYQ JBEADKCQ VMSIT KCUCP LC BLIC RM TM. VABR CIQC TDT AC DKEISTC DK RADQ IDQR?

⬇️ Comment with your guesses

#AlanTuring #Cypher #CrackTheCode #AI #DataScience by @theturinginst

62

1

4 months ago

💡 Want to apply your data skills to real-world challenges?

Apply to join a Data Study Group at the Alan Turing Institute
📍 Week one: Part-time, online
📍 Week two: Full-time, the Alan Turing Institute (@britishlibrary), London

The Turing’s Data Study Groups (DSGs) are intensive collaborative hackathons, uniting organisations from industry, government and the third sector with skilled, multi-disciplinary researchers.

Small teams of carefully selected researchers work together to develop data science solutions to real-world problems which they present at the end of the event and record in a published report.

➡️ Find out more details and FAQs at the link in our bio: @theturinginst

#AI #ArtificialIntelligence #DataScience #DataStudyGroup #Hackathon #Innovation #Research by @theturinginst

9

0

4 months ago

📡 Cleared for Takeoff: Navigating the Future of AI Air Traffic Control

🎙️ Listen to the latest episode of the Turing Podcast to explore the hidden chessboard at the heart of air traffic control and the future of AI in helping to manage our airspace.

Hosted by Amelia Jabry, we’re joined by Ben Carvell, an aviation researcher leading AI agent design on Project Bluebird and Andrew Pace, a London terminal air traffic control officer ATCO).

Project Bluebird is a multi-year partnership between the Alan Turing Institute, @natsaero and the @uniofexeter covering three themes:
✈️ building a high-fidelity digital twin
✈️ developing virtual controller agents
✈️ addressing trust and explainability

Together we explore the challenges and opportunities of building an AI system to assist ATCOs in one of the most safety-critical industries in the world.

🎧 Listen now – link in bio: @theturinginst

#AIAssurance #AirTrafficControl #Collaboration #DigitalTwin #ProjectBluebird by @theturinginst

8

0

4 months ago

Secure your place for the next Turing Lecture with Dr Sasha
Luccioni

Making AI (truly) sustainable – from environmental costs to
social impacts
📅 27 May 2026, 18:00–21:30
📍 Piggott Theatre, @britishlibrary

AI models can transform society – improving efficiency, reducing wasted resources and accelerating innovation. Behind every model trained, every query processed and every dataset stored, though, lies a complex web of environmental, economic and social impacts.

Join us for an evening of insight and discussion where Sasha will lead us through solid, evidence-based steps the AI ecosystem can take towards true sustainability.

Register now – link in bio (@theturinginst)

#ResilientFutures #SustainableAI #TuringLectures by @theturinginst

26

0

5 months ago