Your round-up of the latest, greatest data stories
Hello ODI Supporter,
Last week, we looked at the use of AI in modern warfare. This week, experts warned that data centres have become a new target in conflicts. Iran attacked three data centres operated by Amazon Web Services (AWS) in their retaliatory strikes, two in the United Arab Emirates and one in Bahrain, knocking out facilities and disrupting banking payments, delivery apps and enterprise software in the Middle East. Vincent Boulanin, director of the governance of AI programme at the Stockholm International Peace Research Institute, said, “Data centres are a critical building block of AI capabilities at the national level. From that perspective, data centres can be considered a very critical infrastructure.” These are believed to be the first targeted attacks on datacentres, and while they have caused civilian disruption, it is believed AWS data centres in the region are used for military and intelligence activities. Investment by US tech companies in data centres in the Middle East ramped up in the Biden administration, and was seen as more cost-effective than upgrading facilities on home soil. Now that’s not looking like such a good idea. The attacks have also accelerated plans for six competing projects backed by nations in the region to build overland data routes to Europe as an alternative to undersea cables, which could also be vulnerable to attack. Not so long ago, our research team looked at the vulnerabilities of the cloud, satellites and undersea cables, so take a look at those. Elsewhere, a missile strike on an Iranian girls’ school that is believed to have killed 150 children could have been the result of outdated US targeting data. A US military investigation is still ongoing.
A new study has shown that breast cancer detection could be improved by more than 10% by using an AI tool. A project by NHS Grampian in Scotland was evaluated by the University of Aberdeen. The AI tool, called Mia, can flag small and hard-to-see areas of concern on mammogram scans that can be missed by the human eye. The tool improved detection, while also reducing staff workload and the time it took to notify women of their results. The findings were published in the Nature Cancer journal this week. Earlier detection means that treatment can take place earlier, increasing the likelihood of success.
There is still time to sign up for the hackathon we are running with the Financial Reporting Council (FRC) to accelerate the use of UK structured data. We’re looking for data users, FinTechs, preparers, and subject-matter experts to join us in prototyping solutions and testing the limits of current data availability. It takes place on Friday 27th March, 9:30 am to 5 pm at the FRC offices in London (13th Floor, 1 Harbour Exchange Square, London, E14 9GE). If you’re interested, sign up now.
Join us on Monday 30 March, 12:00-13:00 BST for the latest instalment in our Data Ethics Professionals webinar series, in which we’ll look at key learnings for organisations on embedding data ethics. Tickets are available now. On Thursday 26 March, 16:00-17:00 we’ll be looking at data portals and what their designers and maintainers need to consider as the global digital ecosystem shifts. The next edition of Solid World will look at modelling, analysing and sharing research data, and take place on Monday 23 March, 16:00-17:00. Tickets are available now.
And finally… the lesser spotted woodpecker has been detected for the first time in three decades in the UK, thanks to AI. A team from the University of Sussex set up 37 monitoring stations across Sussex, taking recordings every 10 minutes over the course of a year. They were then able to use AI to identify the woodpecker’s call from more than 1,300 hours of field recordings. The birds are unable to make their nests in living trees, but now their presence has been detected, the woodland where they live can be managed to provide them with the habitat they need.
Until next time.
David and Jo
PS: Our friends at Ovation Data have published a blog this week exploring what the National Data Library and AI-ready data standards mean in practical terms. Check it out.
Data centres are the physical buildings that house all the infrastructure required to power banking apps, cloud services, and artificial intelligence platforms.
The tech industry often talks about “the cloud” as though it were something abstract and untouchable. But the cloud runs on data centers, those data centers have an address, and that address can be hit by a drone.
Saudi Arabia, Qatar, and the UAE are financing competing data corridors through Syria, Iraq, and East Africa to bypass the two maritime choke points that threaten their digital connectivity.
A strike on an Iranian girls' school that killed scores of children may be the result of U.S. use of outdated targeting data, two sources familiar with the matter told Reuters on Wednesday, providing new details about what would rank among the worst cases of civilian casualties in decades of U.S. conflicts.
As the UK advances its National Data Library, organisations holding legacy archives face rising expectations. Data that was once simply stored now needs to be accessible, well‑governed, and AI-ready.
Reporting to the Director of Research, the Head of Research is responsible for scoping, selling and delivering ODI’s research to support the creation of an open, trustworthy data ecosystem.
Over the last eight years, the OpenActive data standard has allowed physical activity providers to publish information about time-based events, such as instructor-led classes or facility time slots. This has led to over 10 million activities being published monthly.
This report examines the stakes and challenges related to ‘critical data infrastructure’, considering both virtual and material dimensions of the internet.
