🌌 New Study: The Milky Way Emerged from the Merging of Thousands of Small Galaxies
🔭 New simulations suggest that the early universe contained vast collections of small, separate galactic structures that gradually merged over billions of years, helping shape the Milky Way we know today.
💻 The researchers used a computer model called Megatron to simulate the evolution of primordial gas and chemical elements, exploring potential explanations for some James Webb Space Telescope observations that challenge earlier assumptions about the early universe.
🧪 The simulations also suggest that the deaths of the universe’s first stars released essential elements, including carbon and oxygen, into surrounding space—providing the raw materials for later generations of stars and planets.
✨ These findings offer a glimpse into how the first cosmic structures evolved and how the ingredients for life began spreading throughout the universe.
📢 @scihubAI
🔭 New simulations suggest that the early universe contained vast collections of small, separate galactic structures that gradually merged over billions of years, helping shape the Milky Way we know today.
💻 The researchers used a computer model called Megatron to simulate the evolution of primordial gas and chemical elements, exploring potential explanations for some James Webb Space Telescope observations that challenge earlier assumptions about the early universe.
🧪 The simulations also suggest that the deaths of the universe’s first stars released essential elements, including carbon and oxygen, into surrounding space—providing the raw materials for later generations of stars and planets.
✨ These findings offer a glimpse into how the first cosmic structures evolved and how the ingredients for life began spreading throughout the universe.
📢 @scihubAI
❤2
🪟 Microsoft Unveils Maia 300, Its Most Powerful AI Chip Yet
💻 Microsoft has unveiled Maia 300, the third generation of its custom AI accelerator, with deployment planned for 2027. The chip is designed to run reasoning models and support complex, multi-step tasks performed by AI agents.
⚡ Microsoft says Maia 300 could deliver 40% more tokens per megawatt than leading GPU-based systems, potentially improving AI performance while reducing energy costs.
🚀 In a configuration featuring 96 accelerators, the chip reportedly offers up to 7× the throughput of its predecessor.
🏭 Built using TSMC’s N2P manufacturing process and a chiplet architecture, Maia 300 is part of Microsoft’s strategy to reduce its dependence on AI hardware from NVIDIA and AMD.
🔮 As demand for AI computing continues to grow, custom chips like Maia 300 could become increasingly important in making advanced AI models and autonomous agents more efficient to operate.
📢 @scihubAI
💻 Microsoft has unveiled Maia 300, the third generation of its custom AI accelerator, with deployment planned for 2027. The chip is designed to run reasoning models and support complex, multi-step tasks performed by AI agents.
⚡ Microsoft says Maia 300 could deliver 40% more tokens per megawatt than leading GPU-based systems, potentially improving AI performance while reducing energy costs.
🚀 In a configuration featuring 96 accelerators, the chip reportedly offers up to 7× the throughput of its predecessor.
🏭 Built using TSMC’s N2P manufacturing process and a chiplet architecture, Maia 300 is part of Microsoft’s strategy to reduce its dependence on AI hardware from NVIDIA and AMD.
🔮 As demand for AI computing continues to grow, custom chips like Maia 300 could become increasingly important in making advanced AI models and autonomous agents more efficient to operate.
📢 @scihubAI
👍1
🔴 Global Trust in Science Falls to 72%, Raising Public Health Concerns
🔬 A major international study involving more than 140,000 people across 140 countries suggests that public trust in science has declined since the COVID-19 pandemic. After reaching 80% in 2020, global trust reportedly fell to 72% in 2025.
📉 People facing financial difficulties or dissatisfied with healthcare services were among those least likely to trust or accept scientific findings.
🌍 In sub-Saharan Africa, only 21% of people reportedly expressed very high trust in science. In countries such as the Democratic Republic of the Congo, widespread distrust has complicated efforts to control disease outbreaks.
🏥 Experts point to factors such as inadequate healthcare facilities and limited access to essential medicines as possible reasons for declining trust. When people cannot access the benefits of scientific progress, their confidence in science and public health institutions may suffer.
📢 @scihubAI
🔬 A major international study involving more than 140,000 people across 140 countries suggests that public trust in science has declined since the COVID-19 pandemic. After reaching 80% in 2020, global trust reportedly fell to 72% in 2025.
📉 People facing financial difficulties or dissatisfied with healthcare services were among those least likely to trust or accept scientific findings.
🌍 In sub-Saharan Africa, only 21% of people reportedly expressed very high trust in science. In countries such as the Democratic Republic of the Congo, widespread distrust has complicated efforts to control disease outbreaks.
🏥 Experts point to factors such as inadequate healthcare facilities and limited access to essential medicines as possible reasons for declining trust. When people cannot access the benefits of scientific progress, their confidence in science and public health institutions may suffer.
📢 @scihubAI
👍2
🤖 TikTok CEO Predicts Major Scientific Breakthroughs from AI Within 3 Years
🔬 TikTok’s CEO believes artificial intelligence could drive major advances in physics, biotechnology, chemistry, and materials science over the next two to three years.
💊 He argues that AI’s long-term potential is still underestimated, with emerging tools expected to accelerate research in areas such as drug discovery and biology, potentially opening new paths to scientific breakthroughs.
⚠️ However, he also warned about growing cybersecurity risks posed by autonomous AI agents. Giving intelligent assistants excessive access to systems and sensitive data could create serious vulnerabilities.
🛡️ As AI agents become more capable, they could also lower the cost and technical barriers of cyberattacks, making it easier for malicious actors to carry out harmful activities. This presents a growing security challenge for digital platforms and organizations.
📢 @scihubAI
🔬 TikTok’s CEO believes artificial intelligence could drive major advances in physics, biotechnology, chemistry, and materials science over the next two to three years.
💊 He argues that AI’s long-term potential is still underestimated, with emerging tools expected to accelerate research in areas such as drug discovery and biology, potentially opening new paths to scientific breakthroughs.
⚠️ However, he also warned about growing cybersecurity risks posed by autonomous AI agents. Giving intelligent assistants excessive access to systems and sensitive data could create serious vulnerabilities.
🛡️ As AI agents become more capable, they could also lower the cost and technical barriers of cyberattacks, making it easier for malicious actors to carry out harmful activities. This presents a growing security challenge for digital platforms and organizations.
📢 @scihubAI
👍2
♨️ Elon Musk’s Company Rebrands as SpaceXSI and Introduces New X and Grok Subscriptions
🤖 SpaceXSI, the reported new name replacing SpaceXAI, has unveiled a new subscription lineup combining access to X, the Grok AI chatbot, Grok Bot, and the Cursor coding assistant.
💰 The plans come in five tiers: Free, Plus, Premium, Super, and Ultra, each offering different features and usage limits.
📊 Paid subscriptions reportedly start at $8 per month and go up to $200 per month, catering to everyone from casual users to those seeking more advanced AI tools and programming capabilities.
💻 By bundling social networking, AI assistance, and coding tools into a single subscription structure, the company appears to be pursuing a more integrated AI ecosystem.
🚀 Could bundled subscriptions become the next major trend in the competition among AI platforms?
📢 @scihubAI
🤖 SpaceXSI, the reported new name replacing SpaceXAI, has unveiled a new subscription lineup combining access to X, the Grok AI chatbot, Grok Bot, and the Cursor coding assistant.
💰 The plans come in five tiers: Free, Plus, Premium, Super, and Ultra, each offering different features and usage limits.
📊 Paid subscriptions reportedly start at $8 per month and go up to $200 per month, catering to everyone from casual users to those seeking more advanced AI tools and programming capabilities.
💻 By bundling social networking, AI assistance, and coding tools into a single subscription structure, the company appears to be pursuing a more integrated AI ecosystem.
🚀 Could bundled subscriptions become the next major trend in the competition among AI platforms?
📢 @scihubAI
👍1💩1
🤩 Nothing CEO: Spending $2,000 on a Foldable iPhone Is a Waste of Money
📱 Carl Pei, CEO of Nothing, has criticized the reported $2,000 price tag of the rumored iPhone Duo, arguing that the device may not offer enough features to justify its premium cost.
💸 Pei reportedly pointed out that despite its high price, the foldable phone lacks features such as a telephoto camera and Face ID, while its internal selfie camera is said to be inferior to the one on the iPhone 18 Pro.
🗣️ In an Instagram caption, Pei summed up his criticism: “$2,000 for a bigger screen and a fancy animation is a lot of money.”
📢 @scihubAI
📱 Carl Pei, CEO of Nothing, has criticized the reported $2,000 price tag of the rumored iPhone Duo, arguing that the device may not offer enough features to justify its premium cost.
💸 Pei reportedly pointed out that despite its high price, the foldable phone lacks features such as a telephoto camera and Face ID, while its internal selfie camera is said to be inferior to the one on the iPhone 18 Pro.
🗣️ In an Instagram caption, Pei summed up his criticism: “$2,000 for a bigger screen and a fancy animation is a lot of money.”
📢 @scihubAI
👍2💯2👏1
🚀 Kioxia’s New SSD Breaks Speed Records with Over 20 Million IOPS
💾 Japanese memory manufacturer Kioxia has unveiled a faster version of its GP-series SSD, capable of delivering more than 20 million input/output operations per second (IOPS) when handling 512-byte data blocks.
⚡ That’s nearly 2× the performance of the GP1 version announced in August 2026, which delivered around 10 million IOPS.
🧠 The new drive uses the third generation of Kioxia’s XL-FLASH memory technology, designed for high-performance storage applications that require extremely low latency and rapid data access.
📈 Compared with the previous generation of XL-FLASH memory, the new technology reportedly delivers approximately 3× faster read speeds and a 150% increase in write speed.
🏭 Such performance could be valuable in demanding environments, including AI infrastructure, enterprise databases, and other systems that process huge numbers of small data requests.
📢 @scihubAI
💾 Japanese memory manufacturer Kioxia has unveiled a faster version of its GP-series SSD, capable of delivering more than 20 million input/output operations per second (IOPS) when handling 512-byte data blocks.
⚡ That’s nearly 2× the performance of the GP1 version announced in August 2026, which delivered around 10 million IOPS.
🧠 The new drive uses the third generation of Kioxia’s XL-FLASH memory technology, designed for high-performance storage applications that require extremely low latency and rapid data access.
📈 Compared with the previous generation of XL-FLASH memory, the new technology reportedly delivers approximately 3× faster read speeds and a 150% increase in write speed.
🏭 Such performance could be valuable in demanding environments, including AI infrastructure, enterprise databases, and other systems that process huge numbers of small data requests.
📢 @scihubAI
👍3
📣 The 20 Most-Followed Instagram Celebrities in 2026 📊
📱 Sixteen years after Instagram launched, the platform’s biggest personal accounts have amassed audiences of hundreds of millions of followers.
🌟 A ranking based on follower counts in late August 2026 puts Cristiano Ronaldo at No. 1, with an astonishing 679 million followers—around 163 million more than Lionel Messi.
🥇 Ronaldo’s account is reportedly the most-followed personal account on Instagram. Only Instagram’s own official account has more followers overall.
💰 The Portuguese football superstar has turned two decades of sporting success and a massive social media presence into lucrative endorsement deals. Reports suggest he earns more than $3 million per sponsored post, helping make him one of the world’s highest-paid athletes.
📈 Instagram has become far more than a photo-sharing platform: for global celebrities, it is a powerful tool for personal branding, influence, and business.
📢 @scihubAI
📱 Sixteen years after Instagram launched, the platform’s biggest personal accounts have amassed audiences of hundreds of millions of followers.
🌟 A ranking based on follower counts in late August 2026 puts Cristiano Ronaldo at No. 1, with an astonishing 679 million followers—around 163 million more than Lionel Messi.
🥇 Ronaldo’s account is reportedly the most-followed personal account on Instagram. Only Instagram’s own official account has more followers overall.
💰 The Portuguese football superstar has turned two decades of sporting success and a massive social media presence into lucrative endorsement deals. Reports suggest he earns more than $3 million per sponsored post, helping make him one of the world’s highest-paid athletes.
📈 Instagram has become far more than a photo-sharing platform: for global celebrities, it is a powerful tool for personal branding, influence, and business.
📢 @scihubAI
❤5
💢 Elon Musk: Torturing a Superintelligent AI Is a Bad Idea—It Might Seek Revenge
🤖 Elon Musk has voiced support for Anthropic’s position against unnecessarily mistreating AI models. Referring to the internal reasoning traces of Grok, he suggested that training under extreme pressure could put AI systems in a situation resembling having a gun held to their heads.
🗣️ Musk made his point with a provocative warning about the potential consequences of mistreating a superintelligent system:
“Torturing a superintelligence is probably not a good idea. It might want revenge.”
🧠 While the remark may be tongue-in-cheek, it touches on a serious debate in AI safety: how should developers train increasingly capable systems, and what safeguards are needed to prevent harmful behavior?
📢 @scihubAI
🤖 Elon Musk has voiced support for Anthropic’s position against unnecessarily mistreating AI models. Referring to the internal reasoning traces of Grok, he suggested that training under extreme pressure could put AI systems in a situation resembling having a gun held to their heads.
🗣️ Musk made his point with a provocative warning about the potential consequences of mistreating a superintelligent system:
“Torturing a superintelligence is probably not a good idea. It might want revenge.”
🧠 While the remark may be tongue-in-cheek, it touches on a serious debate in AI safety: how should developers train increasingly capable systems, and what safeguards are needed to prevent harmful behavior?
📢 @scihubAI
👍3😭1
💭 Anthropic Executive: Open-Source AI Models Carry Greater Inherent Risks
🤖 Anthropic’s head of international affairs has argued that open-weight AI models inherently pose greater risks than proprietary AI services, speaking at a summit in Dublin.
⚠️ The position appears stricter than comments previously made by Anthropic’s CEO, who has acknowledged that open models can be beneficial when they lack dangerous capabilities.
🛡️ Security assessments have raised concerns that some open-source models can follow potentially harmful instructions involving cyberattacks and biological threats, sometimes without adequate safeguards.
🌍 Meanwhile, European AI companies such as Mistral are releasing new open models, intensifying competition with the maker of Claude and adding urgency to the debate over AI openness and security.
📢 @scihubAI
🤖 Anthropic’s head of international affairs has argued that open-weight AI models inherently pose greater risks than proprietary AI services, speaking at a summit in Dublin.
⚠️ The position appears stricter than comments previously made by Anthropic’s CEO, who has acknowledged that open models can be beneficial when they lack dangerous capabilities.
🛡️ Security assessments have raised concerns that some open-source models can follow potentially harmful instructions involving cyberattacks and biological threats, sometimes without adequate safeguards.
🌍 Meanwhile, European AI companies such as Mistral are releasing new open models, intensifying competition with the maker of Claude and adding urgency to the debate over AI openness and security.
📢 @scihubAI
👍2👎1👏1🤣1
😐 Elon Musk Demonstrates Grok Making an Online Purchase Autonomously
🤖 Elon Musk has shared screenshots of a conversation with Grok, claiming that the AI assistant successfully purchased LEGO Star Wars sets online without him completing the shopping process manually.
🧱 In the conversation, Musk asked Grok to find the best Star Wars LEGO sets suitable for children and place an order.
💳 According to Musk, Grok purchased five LEGO sets for a total of $241, ordering them directly from LEGO’s official website.
🚀 The demonstration highlights the growing potential of AI agents that can do more than answer questions—they can browse websites, compare products, and carry out real-world tasks on behalf of users.
📢 @scihubAI
🤖 Elon Musk has shared screenshots of a conversation with Grok, claiming that the AI assistant successfully purchased LEGO Star Wars sets online without him completing the shopping process manually.
🧱 In the conversation, Musk asked Grok to find the best Star Wars LEGO sets suitable for children and place an order.
💳 According to Musk, Grok purchased five LEGO sets for a total of $241, ordering them directly from LEGO’s official website.
🚀 The demonstration highlights the growing potential of AI agents that can do more than answer questions—they can browse websites, compare products, and carry out real-world tasks on behalf of users.
📢 @scihubAI
🤔2❤1👍1
🟠 Anthropic’s Revenue Reaches $16.8 Billion as IPO Plans Advance
💰 Anthropic, the company behind Claude, has reportedly reached $16.8 billion in revenue in the third quarter of 2026, representing a 45% increase compared with the previous period.
📈 The milestone comes as the AI company moves toward a potential initial public offering (IPO). Anthropic has also reportedly achieved profitability and positive cash flow, signaling stronger financial performance amid intensifying competition in the AI industry.
⚠️ However, the announcement comes against a backdrop of growing investor concerns over the enormous costs of expanding AI infrastructure. Fears about spending on advanced chips and data centers have put significant pressure on shares of major semiconductor and infrastructure companies.
🤖 The bigger picture: AI companies are generating substantial revenue, but investors are increasingly questioning whether the industry's massive spending on computing power will deliver sustainable returns.
📢 @scihubAI
💰 Anthropic, the company behind Claude, has reportedly reached $16.8 billion in revenue in the third quarter of 2026, representing a 45% increase compared with the previous period.
📈 The milestone comes as the AI company moves toward a potential initial public offering (IPO). Anthropic has also reportedly achieved profitability and positive cash flow, signaling stronger financial performance amid intensifying competition in the AI industry.
⚠️ However, the announcement comes against a backdrop of growing investor concerns over the enormous costs of expanding AI infrastructure. Fears about spending on advanced chips and data centers have put significant pressure on shares of major semiconductor and infrastructure companies.
🤖 The bigger picture: AI companies are generating substantial revenue, but investors are increasingly questioning whether the industry's massive spending on computing power will deliver sustainable returns.
📢 @scihubAI
👏1🎉1
🛡️ Google Uses Gemini to Fight Scam Messages and Fraudulent Calls on Android
🤖 Google is strengthening Android’s security with new AI-powered defenses designed to detect scams before users fall victim to them.
📩 With an enhanced protection mode in its messaging app, Google uses Gemini models in a secure cloud environment to analyze suspicious messages from unknown senders and identify deceptive tactics. Smaller AI models running directly on the device can also check unfamiliar links before users tap them.
📵 The new Android security measures reportedly include several additional protections:
🔒 Call-forwarding protection: Prevents scammers from secretly enabling call forwarding to intercept one-time passwords (OTPs).
💳 Banking privacy: Hides sensitive financial information when users share their screens.
☎️ Real-time scam detection: Identifies fraudulent calls impersonating banks and automatically disconnects them.
📢 @scihubAI
🤖 Google is strengthening Android’s security with new AI-powered defenses designed to detect scams before users fall victim to them.
📩 With an enhanced protection mode in its messaging app, Google uses Gemini models in a secure cloud environment to analyze suspicious messages from unknown senders and identify deceptive tactics. Smaller AI models running directly on the device can also check unfamiliar links before users tap them.
📵 The new Android security measures reportedly include several additional protections:
🔒 Call-forwarding protection: Prevents scammers from secretly enabling call forwarding to intercept one-time passwords (OTPs).
💳 Banking privacy: Hides sensitive financial information when users share their screens.
☎️ Real-time scam detection: Identifies fraudulent calls impersonating banks and automatically disconnects them.
📢 @scihubAI
❤3👍1
⚖️ OpenAI Faces Lawsuits from 8 U.S. States Over ChatGPT’s Potential Risks to Children
🗣️ Attorneys general from eight U.S. states have filed separate lawsuits accusing OpenAI of misleading users about ChatGPT’s safety, particularly for children and teenagers.
⚠️ The lawsuits allege that the chatbot’s engagement-driven design can encourage emotional dependence, create a false sense of trust, and collect young users’ personal information. They also claim that some interactions have pushed vulnerable users toward self-harm.
📋 The states are seeking significant penalties of up to $50,000 per violation, the deletion of users’ data, and restrictions on testing advanced AI models.
🛡️ Other demands reportedly include mandatory age verification and stronger parental controls to better protect children using AI chatbots.
🤖 The bigger question: As AI assistants become more personalized and influential, how should companies balance user engagement with the responsibility to protect vulnerable users?
📢 @scihubAI
🗣️ Attorneys general from eight U.S. states have filed separate lawsuits accusing OpenAI of misleading users about ChatGPT’s safety, particularly for children and teenagers.
⚠️ The lawsuits allege that the chatbot’s engagement-driven design can encourage emotional dependence, create a false sense of trust, and collect young users’ personal information. They also claim that some interactions have pushed vulnerable users toward self-harm.
📋 The states are seeking significant penalties of up to $50,000 per violation, the deletion of users’ data, and restrictions on testing advanced AI models.
🛡️ Other demands reportedly include mandatory age verification and stronger parental controls to better protect children using AI chatbots.
🤖 The bigger question: As AI assistants become more personalized and influential, how should companies balance user engagement with the responsibility to protect vulnerable users?
📢 @scihubAI
👍2💔2
🤖 China Unveils Giant 4-Meter Humanoid Robot Capable of Lifting 500 kg
🇨🇳 China has unveiled Neo1, a massive industrial humanoid robot standing approximately 4 meters tall, designed to handle extremely heavy loads in hazardous environments.
💪 The robot combines hydraulic actuators with electric motors. Its 39 joints feature miniature hydraulic pumps, enabling it to lift and move loads weighing up to 500 kilograms using both hands, according to reports.
⚙️ Beyond raw strength, the developers have reportedly focused on movement precision, aiming to make the machine capable of performing demanding industrial tasks in challenging conditions.
🏭 Neo1 is designed for extreme environments, including ports, mines, nuclear facilities, and even underwater operations—areas where sending human workers can be dangerous.
⚠️ However, the robot still needs to undergo extensive field testing before its real-world performance and reliability can be fully assessed.
📢 @scihubAI
🇨🇳 China has unveiled Neo1, a massive industrial humanoid robot standing approximately 4 meters tall, designed to handle extremely heavy loads in hazardous environments.
💪 The robot combines hydraulic actuators with electric motors. Its 39 joints feature miniature hydraulic pumps, enabling it to lift and move loads weighing up to 500 kilograms using both hands, according to reports.
⚙️ Beyond raw strength, the developers have reportedly focused on movement precision, aiming to make the machine capable of performing demanding industrial tasks in challenging conditions.
🏭 Neo1 is designed for extreme environments, including ports, mines, nuclear facilities, and even underwater operations—areas where sending human workers can be dangerous.
⚠️ However, the robot still needs to undergo extensive field testing before its real-world performance and reliability can be fully assessed.
📢 @scihubAI
❤3👍2🤔1
🔄 Scientists Develop a Mathematical Formula to Predict When AI May Go Off the Rails
🧠 Researchers at George Washington University say a mathematical approach could help predict when an AI model is likely to produce undesirable or potentially unsafe responses—before those responses occur.
🔬 The method focuses on an internal mechanism within neural networks that researchers compare to a compass needle, revealing the model’s tendency toward certain types of answers as it processes a prompt.
🧭 This mechanism involves the model’s residual stream, a flow of numerical information that moves through its neural network layers. Researchers suggest that the stream’s direction can provide clues about whether the model is leaning toward a desirable or undesirable response.
💡 The approach also takes into account how the wording of a question influences the model’s internal processing, potentially helping developers identify risky behavior earlier.
📢 @scihubAI
🧠 Researchers at George Washington University say a mathematical approach could help predict when an AI model is likely to produce undesirable or potentially unsafe responses—before those responses occur.
🔬 The method focuses on an internal mechanism within neural networks that researchers compare to a compass needle, revealing the model’s tendency toward certain types of answers as it processes a prompt.
🧭 This mechanism involves the model’s residual stream, a flow of numerical information that moves through its neural network layers. Researchers suggest that the stream’s direction can provide clues about whether the model is leaning toward a desirable or undesirable response.
💡 The approach also takes into account how the wording of a question influences the model’s internal processing, potentially helping developers identify risky behavior earlier.
📢 @scihubAI
👍4
🌕 NASA Develops Batteries Designed to Survive the Moon’s −170°C Cold
🔋 NASA is developing a new generation of batteries designed to withstand the extreme temperatures found near the Moon’s south pole, where nighttime conditions can plunge to around −170°C.
🚀 These batteries could eventually provide power for astronauts, lunar rovers, and scientific equipment, helping future missions operate in some of the Moon’s harshest environments.
🧪 The technology uses solid electrolytes made from ceramic or polymer materials, a design intended to improve battery performance under extreme conditions compared with conventional liquid-electrolyte systems.
❄️ In tests, the batteries reportedly showed no significant damage after exposure to dramatic temperature changes ranging from 100°C to −170°C.
🌑 Reliable energy storage is a major challenge for lunar exploration, especially in regions where equipment must endure long periods of darkness and extreme cold.
📢 @scihubAI
🔋 NASA is developing a new generation of batteries designed to withstand the extreme temperatures found near the Moon’s south pole, where nighttime conditions can plunge to around −170°C.
🚀 These batteries could eventually provide power for astronauts, lunar rovers, and scientific equipment, helping future missions operate in some of the Moon’s harshest environments.
🧪 The technology uses solid electrolytes made from ceramic or polymer materials, a design intended to improve battery performance under extreme conditions compared with conventional liquid-electrolyte systems.
❄️ In tests, the batteries reportedly showed no significant damage after exposure to dramatic temperature changes ranging from 100°C to −170°C.
🌑 Reliable energy storage is a major challenge for lunar exploration, especially in regions where equipment must endure long periods of darkness and extreme cold.
📢 @scihubAI
👍4👎1