Anyone who speculates on likely events ahead of time and prepares accordingly can react quicker to new developments. What practically every person does every day, consciously or unconsciously, is also used by modern computer processors to speed up the execution of programs. They have so-called speculative technologies which allow them to execute instructions on reserve that experience suggests are likely to come next. Anticipating individual computing steps accelerates the overall processing of information.
However, what boosts computer performance in normal operation can also open up a backdoor for hackers, as recent research by computer scientists from the Computer Security Group (COMSEC) at the Department of Information Technology and Electrical Engineering at ETH Zurich shows.
The computer scientists have discovered a new class of vulnerabilities that can be exploited to misuse the prediction calculations of the CPU (central processing unit) in order to gain unauthorized access to information from other processor users. They will present their paper at the 34th USENIX Security Symposium (USENIX 2025), to be held August 13–15, 2025, in Seattle.
In a new Nature Communications study, researchers have developed an in-memory ferroelectric differentiator capable of performing calculations directly in the memory without requiring a separate processor.
The proposed differentiator promises energy efficiency, especially for edge devices like smartphones, autonomous vehicles, and security cameras.
Traditional approaches to tasks like image processing and motion detection involve multi-step energy-intensive processes. This begins with recording data, which is transmitted to a memory unit, which further transmits the data to a microcontroller unit to perform differential operations.
Please see my latest Security & Tech Insights newsletter. Thanks and have a great weekend!
Link.
Dear Friends & Colleagues, please refer to the latest issue of the Security & Tech Insights newsletter. In this issue, several articles highlight emerging tech trends for 2025. Some of these topics were also selected by Forrester’s research on emerging technologies in 2025, which highlights tech that will help drive AI-led innovation while enabling long-term resilience. Thanks for reading and stay safe! Chuck Brooks.
Countries in the Global South risk being left out of the quantum revolution — along with its economic, technological and security benefits — due to growing export controls, siloed research initiatives and national security concerns, a new policy analysis argues.
In the first of a series of articles on quantum technologies published by the policy journal Just Securit y, researchers Michael Karanicolas, of Dalhousie University, and Alessia Zornetta, of UCLA Law, examine how the geopolitics of emerging quantum technologies are replicating long-standing patterns of technological exclusion. The authors argue that absent meaningful interventions, quantum could become another engine of global inequality, one that threatens to lock poorer nations out of the next era of technological and economic development.
The authors trace the roots of this divide to export control regimes that are quickly expanding in response to the strategic potential of quantum systems. Since 2020, governments in the U.S., EU and China have implemented targeted restrictions on quantum-enabling hardware, software, and communications systems.
Robotics is now revolutionizing numerous industry sectors through the integration of artificial intelligence, machine learning, and reinforcement learning, as well as advances in computer vision that empower robots to make complicated judgments.
Industrial automation in factories and warehouses has been the main emphasis of robotics for many years because of its efficiency and affordability. These settings are usually regulated, organized, and predictable. Consequently, industries like manufacturing, agriculture, warehouse operations, healthcare, and security have utilized robotics to automate mundane programmable tasks.
Robotics in those and many other industries are becoming more refined and capable with the contributions of new material sciences, and artificial intelligence tools. It now appears that with those advances, we are at the precipice of building functional, dexterous, and autonomous humanoid robots that were once the topic of futurist writing.
This month’s AI news covers major breakthroughs, including humanoid robots that run and think faster than humans, and China deploying real robotic AI police on the streets. We also explore DeepMind accidentally breaking its own AI, Microsoft building its most efficient model yet, and Meta releasing a two-trillion-parameter AI called Llama 4. Plus, DeepSeek’s new self-learning AI, China’s ultra-fast AI agents, and next-gen video generators that look more real than reality are changing the game.
A humanoid robot that runs and thinks faster than humans
China’s real AI-powered police robots now patrolling streets
DeepSeek’s new self-learning AI rivaling top-tier models
DeepMind breaks its own AI with a single prompt
Microsoft accidentally creates its most efficient AI yet
Meta releases a massive two-trillion-parameter model
China unveils ultra-fast AI agents and hyper-real video generators
🎥 What You’ll See:
Advanced humanoid AI in action
Robotic cops deployed across Chinese cities
Self-improving AI models that beat OpenAI in key areas
DeepMind’s AI failure revealing system vulnerabilities
Meta’s Llama 4 shaking up the AI model race
China’s AI creating videos that look better than real life
📊 Why It Matters: From real-world AI deployments to record-breaking models, this month shows how fast AI is evolving—reshaping robotics, security, video generation, and self-learning systems in ways we’ve never seen before. #ai #openai #deepseek. Get the best AI news without the noise 👉 https://airevolutionx.beehiiv.com/
🔍 What’s Inside: A humanoid robot that runs and thinks faster than humans. China’s real AI-powered police robots now patrolling streets. DeepSeek’s new self-learning AI rivaling top-tier models. DeepMind breaks its own AI with a single prompt. Microsoft accidentally creates its most efficient AI yet. Meta releases a massive two-trillion-parameter model. China unveils ultra-fast AI agents and hyper-real video generators.
🎥 What You’ll See: Advanced humanoid AI in action. Robotic cops deployed across Chinese cities. Self-improving AI models that beat OpenAI in key areas. DeepMind’s AI failure revealing system vulnerabilities. Meta’s Llama 4 shaking up the AI model race. China’s AI creating videos that look better than real life.
📊 Why It Matters: From real-world AI deployments to record-breaking models, this month shows how fast AI is evolving—reshaping robotics, security, video generation, and self-learning systems in ways we’ve never seen before.
Researchers at the University of Rochester and Rochester Institute of Technology recently connected their campuses with an experimental quantum communications network using two optical fibers. In a new paper published in Optica Quantum, scientists describe the Rochester Quantum Network (RoQNET), which uses single photons to transmit information about 11 miles along fiber-optic lines at room temperature using optical wavelengths.
Quantum communications networks have the potential to massively improve the security with which information is transmitted, making messages impossible to clone or intercept without detection. Quantum communication works with quantum bits, or qubits, that can be physically created using atoms, superconductors, and even in defects in materials like diamond. However, photons—individual particles of light—are the best type of qubit for long distance quantum communications.
Photons are appealing for quantum communication in part because they could theoretically be transmitted over existing fiber-optic telecommunications lines that already crisscross the globe. In the future, many types of qubits will likely be utilized because qubit sources, like quantum dots or trapped ions, each have their own advantages for specific applications in quantum computing or different types of quantum sensing.
