AI topics

Ka-band metasurface antennas present a promising solution for satellite communications due to their cost-effectiveness, compact design, and advanced beam-steering capabilities. However, challenges arise from the limitations of satellite resources and the significant atmospheric losses associated with Ka-band frequencies. To address these issues, it is essential for these antennas to achieve wide-angle beam scanning and high antenna gain, which introduces considerable complexity into their design.

Geoffrey Hinton, a pioneer in artificial intelligence, celebrated his recent Nobel Prize in physics by driving a rented car to Google's headquarters in California.

Demis Hassabis and Geoffrey Hinton have received significant recognition for their contributions to artificial intelligence, culminating in two Nobel prizes. Geoffrey Hinton, alongside John Hopfield, was awarded the Nobel Prize in Physics for their foundational work on artificial neural networks, which form the basis of contemporary AI systems like ChatGPT. Meanwhile, the Chemistry prize was awarded to Demis Hassabis and John Jumper for their work on AlphaFold, a program developed by Google DeepMind that successfully predicts the structures of all proteins, addressing a long-standing scientific challenge. This recognition highlights a pivotal moment in the evolution of AI technology.

Arne Schepker, CEO of the Berlin-based language learning platform Babbel, is resigning from his position. Co-founder and former CEO Markus Witte will return to lead the company during the search for Schepker's successor. This transition is expected to focus on integrating artificial intelligence into Babbel's future strategies.

The Hong Kong-Zhuhai-Macao Bridge is a significant infrastructure project that connects Hong Kong, Zhuhai, and Macao, extending over 55 kilometers. It incorporates a "bridge-island-tunnel" design, which includes a 6.7-kilometer undersea tunnel and two artificial islands. The total investment for the project amounts to approximately 120 billion yuan.

The 2024 Nobel Prizes in physics and chemistry highlight the significant role of artificial intelligence in advancing scientific discoveries. These awards reflect a transformative shift in research methodologies, prompting speculation about Alfred Nobel's perspective on the integration of AI in contemporary science.

Madeleine Finlay engages with the Guardian science team, comprising Nicola Davis, Ian Sample, and Hannah Devlin, to discuss the recent Nobel Prize announcements. Recognitions include awards for the discovery of microRNA and advancements in protein creation, alongside accolades in artificial intelligence within the fields of physics and chemistry. The discussion highlights the significant contributions and unexpected outcomes associated with this year's Nobel Prizes in science.

China Telecom, a major state-owned telecommunications company, has successfully developed two large language models (LLMs) trained exclusively on domestically-produced chips. This achievement marks a significant advancement in China's ambition for self-reliance in artificial intelligence (AI), particularly amid increasing US restrictions on access to advanced semiconductors. The company’s Institute of AI announced that one of the models, TeleChat2-115B, and another unnamed model were trained using tens of thousands of Chinese-made chips. This development highlights China's capability to independently create LLMs and signifies a new phase of innovation in the field. The unnamed model reportedly consists of one trillion parameters, making it highly complex and powerful. In contrast, the TeleChat2-115B model contains around 100 billion parameters, offering competitive performance akin to mainstream AI platforms. With US export controls limiting access to high-end processors like Nvidia's A100 and H100, Chinese firms have been compelled to in…

AI enhances the detection of cybersecurity threats but also introduces more sophisticated challenges. Research by Keeper Security reveals that many organizations are not adequately prepared for AI-driven threats, despite having implemented related policies. A significant 84% of IT and security leaders believe that AI-enhanced tools have made detecting phishing and smishing attacks more difficult. In response, 81% of organizations have instituted AI usage policies, with 77% of leaders feeling confident in their understanding of AI security best practices. However, over half of the security leaders consider AI-driven attacks the most serious threat to their organizations, and 35% feel unprepared to tackle these specific threats. Key strategies being deployed include data encryption (51%), employee training (45%), and advanced threat detection systems (41%). To effectively combat AI-driven cyber threats, organizations are encouraged to maintain fundamental cybersecurity practices while adopting advanced frameworks like zero trust and Privileged Access Management (PAM). Zero trust emphasizes continuous verification of all users and devices, reducing the risk of unauthorized ac…

Artificial intelligence continues to showcase its revolutionary potential, as evidenced by the recent Nobel Prize in Chemistry awarded to Demis Hassabis and John M. Jumper from Google DeepMind for their pioneering work in applying AI to molecular biology. They received this prestigious recognition for their contributions to protein structure prediction through AI, alongside David Baker, a professor at the University of Washington, who was honored for his innovative computational protein design tools. Hassabis and Jumper are the creators of AlphaFold, a groundbreaking innovation that has significantly accelerated the process of predicting the three-dimensional structures of proteins from amino acid sequences, a task that previously took years. Their latest model, AlphaFold 3, extends this capability to predicting the structures of DNA, RNA, and ligands, essential for drug discovery. The impact of this technology is profound, potentially leading to more effective vaccines, advancements in cancer research, and the development of new materials. AlphaFold has been widely adopted by researchers globally, with over two million users leveraging its capabilities for critical scien…