Welcome to our latest round-up of news from the technology and hosting world. Here’s what we’ve discovered this month.
AI Cracks Superbug Mystery
A microbiology problem that took a team at Imperial College London (UCL) more than ten years to understand has been cracked by a Google-developed AI tool in just two days.
UCL professor, José R Penadés, provided the AI with a brief prompt regarding his unpublished, ten-year study on how superbugs gain resistance to antibiotics and was amazed when it produced the correct hypothesis, as well as several other valid suggestions – one of which was entirely new to the research team and is now under further investigation.
With his own research not yet published, Penadés was initially worried that the tool had accessed his confidential information. However, he was reassured by Google that this was not the case, and that the AI had solved the problem by itself. Calling the tool ‘spectacular’, the professor believes that AI’s speed and capabilities could transform scientific research.
Heritage Disaster Robot
According to Historic England, robotic dogs and drones have the potential to transform disaster response at the UK’s heritage sites. If devastating fires, similar to the one at York Minster in 1984, were to happen today, advanced robo-dogs could be utilised to safely evaluate the damage and even move debris, as they did following the fire at Notre Dame Cathedral in Paris.
The robo-dog Spot, developed by Boston Dynamics, which can regain balance if knocked over, is capable of manoeuvring through debris, carrying scanners that deliver detailed internal assessments of damaged structures. When paired with drones that can reach difficult locations, these technologies provide a quicker way to manage recovery efforts. Importantly, both can be piloted remotely, allowing operators to stay safe while assessing conditions.
The destruction of York Minster’s Rose Window in 1984 took years to fix, and access to it is still challenging now, requiring large scaffolding structures to be erected. Robot and drone technology make it much easier to monitor these hard-to-access parts of historic buildings, making them essential tools for today’s emergency response and maintenance.
Oxford Partners with OpenAI
The University of Oxford has begun a five-year partnership with OpenAI that will provide students and researchers with access to advanced AI tools and funding opportunities. The joint venture aims to improve research, teaching and learning, especially in areas like health and climate science. As part of the programme, a version of ChatGPT specifically designed for education will be accessible to the university. Oxford researchers will also receive early access to OpenAI’s newest models, facilitating complex data analysis across various academic fields.
In addition, the partnership will see some of the Bodleian Library’s collection digitised for worldwide access, fulfilling the library’s goal of preserving and sharing knowledge through innovative methods.
The collaboration is part of the larger NextGenAI initiative, which seeks to connect leading UK and US universities with OpenAI to enhance AI research and prepare institutions for the long-term impact of the technology.
Future Tech Lessons
Secondary school students in Lancashire will soon be able to engage with new technologies, such as AI, cybersecurity and virtual reality. The Lancashire MEGA Hubs programme, a new proposal supported by the county council, aims to improve digital skills and prepare students for careers in the expanding tech industry.
Every secondary school in the area will be able to participate in the programme until the end of the 2026 academic year. Students will take part in workshops on coding and machine learning, and they can join an extracurricular Cyber and Esports Academy.
Funded by a £550,000 grant from the Lancashire Economic Growth & Development Investment Fund and implemented in collaboration with the IN4 Group, the project will also incorporate digital tools into conventional STEM subjects. The programme is part of a broader effort to cultivate a diverse and tech-savvy workforce, offering early exposure to digital skills for all students, regardless of their backgrounds.
Multichannel AI Attacks
Research by SoSafe found that 94% of organisations reported a rise in multichannel cyberattacks last year. These sophisticated attacks use multiple platforms, such as email, SMS and messaging apps, to imitate legitimate communications and trick victims.
Increasingly, these attacks are using AI. One highly convincing fraud scenario discovered by SoSafe involves attackers using deepfake voice calls alongside messages sent via WhatsApp and Microsoft Teams. Cybercriminals take advantage of the interaction between different channels to create a sense of urgency and reduce suspicion, further increasing the risk for businesses.
The complex nature of multichannel attacks increases their believability, making them more difficult to detect, even for trained personnel. Indeed, only 25% of companies felt confident in their ability to identify them. With AI-based attacks predicted to become more frequent and severe, businesses are encouraged to reassess their security measures and training programmes.
Visit our website for more news, blog posts, knowledge base articles and information on our wide range of hosting services.
