Secrets of Intelligent Technology
A lot of industries have been incorporating Artificial Intelligence (AI) into their work over the years, and this has helped increase their degrees of accuracy and replicability, which is a great thing in our technology-driven world. In today’s newsletter, we’ll take a look at how AI and Machine Learning (ML) are impacting other aspects of technology, such as how AI is being used to reduce mechanical friction, increase the chances of transplants, improve interactions between humans and AI, and make the internet safer.
Machine learning is helping to stabilize mechanical friction conditions:
One of the first science principles every kid learns is that friction is the bane of every machine. This resisting force leads to wear and tear and the eventual failure of parts. Well, MIT researchers have made a machine learning algorithm that can stabilize mechanical friction. This could make machines more reliable and efficient. The algorithm uses a neural network to predict friction forces based on sensor data, which is then used to adjust the machine's operating parameters. The researchers tested the algorithm on a variety of machines, including a robotic arm and a drill press, and found that it improved performance and reduced wear and tear on the machines. This change is exciting for the manufacturing industry because it could lead to machines that work better and last longer.
Humans and AI are collaborating in real time:
Automated design systems and 3D printers have made it easier to make things with complicated shapes, like medical devices and parts that hold up skyscrapers. But automated systems alone might not be enough to make designs that are fully optimized for their purpose. To address this, MIT researchers have developed a Human-Informed Topology Optimization system that combines the benefits of automated design with human engineer input. The system starts by setting the requirements, and the human engineer gives continuous feedback on how the work is going. The algorithm incorporates this feedback, producing optimized designs based on various properties. With this technology, AI and human engineers might be able to work together to make designs that are faster and better. It can make a huge impact in disaster-relief areas or war zones where computational power is limited.
AI is being used to select organs for transplantation:
With the help of AI, a new technology called Organ Quality Assessment (OrQA) will change the transplant system in the UK in a big way. OrQA uses deep machine learning, which is similar to facial recognition technology, to judge the quality of organs that people want to donate. The software is meant to help surgeons figure out if organs are healthy enough to be transplanted, which could lead to more people getting transplants that save their lives. Developed by researchers at the University of Bradford, the technology is being supported by the NHS Blood and Transplant Quality in Organ Donation Biobank and the NIHR Blood and Transplant Research Unit. Once it is clinically validated, the tool will be licensed for use in the NHS and could also be marketed worldwide, saving lives and improving the quality of life for those awaiting organ transplants.
AI for surveillance is growing:
AI has made it possible for surveillance technology to not only identify people but also figure out who they are connected to. With the help of AI, co-appearance analysis or correlation analysis software can look at surveillance footage to find people who have been seen together more than once. This makes it possible to find people who might be working together. Even though authoritarian countries like China have already used this technology, companies in the West are also testing these new AI and surveillance applications. Concerns have been raised, though, about the possible invasion of privacy and violation of rights protected by the Constitution. As AI keeps getting better, it will become more important to find a good balance between security and privacy.
DDoS attacks are being foiled faster by AI:
Distributed denial-of-service (DDoS) attacks are a growing threat to organizations, and researchers are exploring ways to improve cybersecurity defenses. One way to deal with DDoS attacks is to use AI to find them and respond to them more quickly and effectively. Researchers at the University of California, Riverside, have made an AI-based tool that can find and stop DDoS attacks in real time. The tool uses algorithms for machine learning to look at network traffic and find patterns that show an attack is happening. Once an attack is found, the tool can change network settings automatically to stop the attack. | 
