The top 10 most exciting technological innovations: how 2024 has delighted humanity.
The past year has been filled with significant technological achievements. As expected, the main theme revolved around artificial intelligence (AI) and the challenges that arise alongside its development. Despite the overwhelming popularity of this topic, scientists, startups, and tech corporations have introduced truly impressive innovations this year.
The potential applications of these technologies are incredibly vast, ranging from addressing physical and mental health issues to data transmission in space and forest conservation.
Traditionally, at the end of the year, EP publishes a selection of the most interesting technological developments that deserve attention.
Text-to-Video Transformation
In 2024, it is hard to surprise anyone with the capabilities of generative AI, which can create texts in various styles based on any request. Just a year ago, experts in the field of AI agreed that the development of large language models would focus on their practical application in specific professions, such as medicine or law.
Another promising direction of development was considered to be the next stage in content creation: generating audio, images, and videos. While audio and image generation became available (albeit with certain inaccuracies) last year, creating videos remained a more complex task for developers.
On February 15, OpenAI introduced a new AI system called Sora, capable of generating realistic videos based on textual queries. On December 9, Sora became available to the general public. The new model allows users to create videos up to 20 seconds long based on text descriptions. Initially, the number of generations will be limited.
During the presentation, OpenAI demonstrated a new feature on the Sora platform - the "Explore" page, which features a feed of videos created by other users using this system. The developers introduced features such as "storyboarding" (creating videos based on a sequence of text queries), "mixing" (which allows combining elements from two scenes within one video), and "remix" (for adjusting generation results through additional text queries).
Although Sora is not the only AI model for video creation, its functionality, attention to detail, and the quality of work from OpenAI engineers make the results as realistic as possible.
The Future of Home Medicine
The French company Withings attracted attention this year at the international CES 2024 exhibition, showcasing its new product - the compact multifunctional device BeamO.
The compact all-in-one device is equipped with a joystick that allows for easy switching between its main functions. The developers promise that the device can perform all measurements in just one to two minutes. The results are automatically transmitted to the Withings Health Mate app on smartphones via Wi-Fi or Bluetooth, where they can be viewed or sent to a doctor.
BeamO exemplifies how new technologies are driving the development of telemedicine, as doctors can monitor key patient metrics in real-time.
The autonomy of the BeamO device lasts eight months, and it can be charged via a USB-C cable. Initially, it was planned to be available for purchase in the summer of 2024 in the US and Europe at a price of $250. However, there is currently no information on the official start of sales on the Withings website. The preliminary sales launch is expected to take place in 2025.
Changing the Walking Culture
According to Kim Yong-jae, CEO of WIRobotics, the company aimed for their exoskeleton to be not just a prototype of a research project but a commercial product. Engineers created a device weighing less than 2 kg, which is convenient to wear and easy to transport. Moreover, it can even be used while driving a car.
The main mission of the team is to change the perception of the familiar act of walking. Unlike traditional exoskeletons designed for the elderly or individuals with mobility impairments, WIM is developed for young users who want to improve their fitness or strengthen their muscles during workouts.
With the help of a special app, WIM users can create personalized walking programs, analyzing data on strength, body balance, and posture. The analysis itself is based on AI technologies.
Developed by former Samsung engineers, the robot has two operating modes: support mode, which reduces energy expenditure during walking by 20%, and training mode, which creates resistance similar to walking in water to strengthen muscles.
5The exoskeleton features a single motor at the waist and thigh extensions with individual drives, and it can analyze and correct the user's gait in real-time.
WIRobotics expects that the stylish and vibrant design will make WIM a popular accessory among Generation Z and millennials. The device is easy to disassemble, simplifying transportation and storage, making it ideal for an active lifestyle. The price of one model is around $2500.
Flying Hydrogen-Electric Taxis
Joby Aviation, a company developing next-generation air transport, successfully conducted a test flight of its first-class hydrogen-electric aircraft S4 eVTOL this year, covering a distance of 842 km. This is three times the range of its battery-powered version.
6The aircraft, often referred to as "flying taxis," are designed to transport passengers over short distances within cities. The successful test flight was a significant event, as it likely marked the first flight in history of a vertical takeoff and landing aircraft using liquid hydrogen as an energy source.
Joby Aviation plans to launch air taxis in 2025 for urban transport in New York and Los Angeles in collaboration with Delta Air Lines and Uber, as well as in Dubai and Abu Dhabi.
The new air taxi is a modified electric aircraft with six rotors, designed for urban environments. Its previous battery-powered version has undergone over 40,000 km of testing and has made numerous flights based at the company in Marina, California, and over New York City.
Revolution in Fighting Forest Fires
This year, Google Research, in collaboration with Earth Fire Alliance and other partners, introduced an innovative satellite project FireSat, aimed primarily at monitoring forest fires. The system will collect data on the size, intensity, and spread of fires, which will aid in combating this global issue.
Wildfires have a significant impact on climate change due to massive carbon emissions. For instance, the fires in Canada in 2023 resulted in emissions of 3 billion tons of CO2.
The FireSat satellite constellation will provide real-time data on fire locations, helping rapid response teams address fires in their early stages. This is especially crucial for countries with limited resources for wildfire management.
FireSat will offer global high-resolution imagery that updates every 20 minutes. It will be capable of detecting fires as small as a few square meters. Using artificial intelligence, the system will analyze images, comparing them with thousands of previous frames of the same location, while also considering weather conditions and other local factors. This will allow for precise fire identification.
In regions with developed infrastructure, such as California, FireSat will complement existing technologies, particularly those from the US Department of Defense and unmanned surveillance systems. In countries and regions where such technologies are lacking, FireSat will represent a true breakthrough.
The first satellite is expected to launch in 2025, and full deployment of the network will take several more years. Once the FireSat project is completed, it will become an essential tool for response teams and scientists, helping to refine firefighting methods and better understand their nature.
Neuromorphic Systems to Reduce AI Costs
In April of this year, Intel introduced Hala Point - the world's largest neuromorphic computer designed to support research in artificial intelligence, inspired by the principles of brain function.
Neuromorphic systems represent a new paradigm in computing, mimicking the structure and functions of the human brain. Unlike traditional computers with centralized processors and separate memory, these systems operate based on parallel information processing, characteristic of the brain's neural networks.
7Hala Point is built on more than a thousand Loihi 2 processors, which include 140,000 neuromorphic processing cores. The system demonstrates high performance with the ability to perform up to 30 quadrillion (a number with 15 zeros) operations per second, as well as energy efficiency at 15 trillion 8-bit operations per second per watt.
The memory bandwidth