AGIBOT has announced a major achievement, rolling out its 10,000th humanoid robot on March 30, 2026, in Shanghai. This makes it one of the first robotics companies in the world to reach such a scale. The milestone reflects a clear shift in the industry, from early-stage testing to large-scale real-world deployment. It also signals that humanoid robots are no longer limited to labs or experiments but are becoming part of everyday business environments.

From Experiment to Real-World Use

This achievement is not just about production numbers. It highlights AGIBOT’s ability to deploy robots at scale across different industries. The company is moving beyond niche use cases and entering mainstream commercial adoption. This marks an important transition where the focus is no longer just on whether the technology works, but on how it delivers real business value. It also shows that embodied AI is becoming practical and usable in real-world settings.

Rapid Production Growth

AGIBOT’s growth has accelerated significantly over time. It took nearly two years to produce its first 1,000 robots. The jump from 1,000 to 5,000 units took about a year. However, scaling from 5,000 to 10,000 units happened in just three months. This fourfold increase in speed highlights major improvements in manufacturing efficiency and supply chain maturity. It also reflects rising demand for humanoid robots across industries.

According to Peng Zhihui, this rapid scaling is the result of a mature supply chain and standardized manufacturing processes. He emphasized that demand for large-scale commercial robotics solutions is increasing quickly. The company is focused on expanding the adoption of embodied AI systems and making them more accessible to businesses worldwide.

Expanding Real-World Applications

A large number of AGIBOT’s robots are already deployed in real-world environments. They are being used in logistics, retail navigation, hospitality services, and education. The company is also expanding into industrial applications, including production line support and manufacturing operations. This shows a clear shift toward practical, day-to-day use of humanoid robots.

Global Reach and Industry Impact

AGIBOT is also seeing strong international demand, with deployments across Europe, North America, Japan, South Korea, Southeast Asia, and the Middle East. The focus is moving from pilot projects to repeat large-scale rollouts. This milestone confirms that humanoid robots are now scalable, commercially viable, and ready to be integrated into everyday business operations, positioning AGIBOT as a leader in the global robotics industry.

What Is AGIBOT?

AGIBOT is a Shanghai-based robotics company focused on humanoid robots powered by embodied AI for industrial and service use. Founded in 2023, it has quickly scaled from prototypes to mass production, crossing 10,000 robots and becoming one of the first to deploy humanoids at this level. Its systems are already used in logistics, retail, hospitality, and manufacturing environments. With rapid production growth and expanding global deployments, AGIBOT is now focused on scaling real-world adoption and making general-purpose robots more practical across industries.

Read More:

• China vs US in Robotics: Nvidia CEO Reveals the Real Power Shift
• ANTHBOT M5 Smart Robot Lawn Mower Gets Massive Discount – Limited Time Amazon Deal
• AI-Powered Robots Begin Real Battlefield Testing

The post Humanoid Robots Are Scaling Fast, And AGIBOT from China Is Leading the Charge appeared first on Gizmochina.

OpenAI is reportedly shutting down its Sora video-generation tool, less than two years after it first drew widespread attention. Sora was introduced in 2024 as a way to create realistic video clips from simple text prompts. Early demos generated a lot of buzz, especially around how quickly the quality seemed to improve.

But that momentum didn’t quite translate into long-term traction. OpenAI has now announced that it plans to step back from video generation and focus more on areas like robotics and so-called “agentic” AI.

That said, the situation around Sora isn’t a clean, immediate shutdown. While OpenAI has indicated that it is discontinuing the app and winding down development, the tool hasn’t disappeared overnight. The web version, for instance, is still accessible for now, and there’s no clearly defined cutoff date yet. It suggests Sora is in a transition phase—no longer a priority, but not fully offline either.

The shift isn’t just about the product itself. OpenAI is also walking away from a major content partnership with The Walt Disney Company, which had been announced only a few months ago. The deal, reportedly worth around $1 billion, would have allowed Sora users to generate videos featuring well-known Disney characters.

According to reports, the agreement hadn’t progressed far enough for any payments to be made, and Disney was caught somewhat off guard by the decision. The company has since said it will continue exploring AI partnerships elsewhere, while keeping a close eye on how its intellectual property is used.

Sora’s commercial performance may have played a role here. Despite the initial excitement, the platform reportedly brought in around $1.4 million in global in-app revenue—a relatively small figure compared to ChatGPT’s $1.9 billion over a similar period. There were also ongoing concerns around moderation, including misuse involving deepfakes, copyrighted content, and non-consensual imagery.

At the same time, competition in the space has been heating up, with new tools emerging from other companies. Combined with legal risks and rising infrastructure costs, continuing Sora may have been harder to justify.

Don’t miss a thing! Join our Telegram community for instant updates and grab our free daily newsletter for the best tech stories!

For more daily updates, please visit our News Section.

(Source)

The post OpenAI discontinuing its AI video generation tool Sora appeared first on Gizmochina.

Key Highlights

• Humanoid robots like Phantom MK-1 are now being tested in real combat environments
• Over 7,000 robot missions completed in a single month show rapid adoption
• Future warfare may rely on coordinated AI robots across land, air, and sea

The use of AI-powered robots in warfare is no longer just an idea; it is becoming a reality. Modern battlefields are now being used to test advanced machines designed to reduce human risk and improve efficiency. These developments show how quickly robotics and artificial intelligence are moving from labs into real-world situations.

Humanoid Robots Enter the Field

One of the most advanced examples is the Phantom MK-1 humanoid robot. It is designed to move like a human and operate in difficult terrains where traditional machines struggle.

The robot stands around 175 cm tall, weighs about 80 kg, and can carry up to 20 kg. It uses cameras and sensors to understand its surroundings and can move at speeds of up to 6 km/h.

These robots are not fully independent. They are being tested to study mobility, performance, and how AI behaves under pressure.

How AI Military Robots Work

Military robots today use a mix of AI and human control. This is called a “human-in-the-loop” system. AI helps with tasks like identifying objects, navigating terrain, and suggesting actions. However, humans still control critical decisions, especially when it comes to using weapons.

Rise of Ground Robots

Humanoid robots are only part of the story. Uncrewed Ground Vehicles (UGVs) are already widely used.

In January 2026 alone, more than 7,000 missions were carried out using robots. These machines mainly handle logistics such as delivering supplies, evacuating injured soldiers, and scouting areas. Most robots are currently used for support tasks rather than direct combat.

Challenges and Future Potential

Despite rapid growth, there are still limitations. Robots face issues like limited battery life, high costs, and difficulty understanding complex situations. There are also concerns about hacking and misuse.

Looking ahead, experts believe future warfare could involve large groups of connected robots working together across land, air, and sea. This shift is not just about warfare; it is a major step forward in robotics and AI. Machines are slowly moving from tools to active partners, shaping the future of technology.

Did You Know? 

The Phantom MK-1 is built by a San Francisco-based startup called Foundation, founded by former military personnel and engineers focused on defense robotics. The company has already secured about $24 million in contracts with the US Army, Navy, and Air Force, making it an official defense partner. 

Beyond this robot, the global race for military robotics is accelerating; countries like the United States, China, Israel, and Russia are actively developing and deploying robotic systems. China has tested armed robot dogs in military drills, while the US has long used systems like PackBot and TALON in combat zones. Even countries like Estonia and Turkey are building advanced unmanned ground and aerial combat systems, showing that the future battlefield is rapidly becoming automated.

Read More:

• US Router Ban Explained: Should You Be Worried?
• Forgot Passwords: Dell’s Latest Mouse Unlocks Your PC
• China’s Military Robot Wolves Revealed: Armed Packs Ready for Urban Warfare

(via)

The post AI-Powered Robots Begin Real Battlefield Testing appeared first on Gizmochina.

• China unveils advanced robotic wolf packs built for urban warfare
• Equipped with micro missiles, grenade launchers, and AI coordination
• Signals rising global race in autonomous military robotics

China has revealed a new generation of robotic “wolf packs,” with first-ever footage showing them operating in simulated urban combat. The development was reported on March 26 by CCTV News, highlighting a major leap in autonomous battlefield technology.

The systems have been developed by China Ordnance Automation Research Institute Co., Ltd. and represent the latest evolution of robotic combat units. Compared to earlier versions, these machines feature a stronger physical build designed to handle harsh environments, along with significantly improved AI “brains” that enable smarter, more autonomous decision-making. Their enhanced coordination also allows them to operate effectively in packs, increasing combat efficiency.

A key highlight is their ability to carry and deploy multiple weapons. These robotic wolves can be equipped with micro missiles and grenade launchers, making them capable of firepower-based target suppression in dense urban warfare scenarios. Their deployment suggests a focus on close-range, high-risk environments where human soldiers face greater danger.

Each unit is assigned a specialized role within the pack. “Shadow” units focus on reconnaissance and scouting, “Bloodstained” units are designed for direct attack, and “Polar” units provide operational support. This structured role system allows for coordinated, team-based tactics similar to human military squads.

Control systems are equally advanced. Operators can guide these machines using voice commands or gesture-based controls via data gloves, enabling real-time human-machine coordination on the battlefield.

Globally, similar technologies are being explored, particularly in the United States. Companies like Boston Dynamics have developed quadruped robots such as Spot, primarily used for surveillance, inspection, and logistics. While the U.S. military has tested armed robotic platforms, it has generally taken a more cautious approach toward deploying fully autonomous combat systems.

The release of this footage signals intensifying competition in military robotics, as nations push toward smarter, more coordinated, and increasingly weaponized autonomous systems.

Read More:

• China vs US in Robotics: Nvidia CEO Reveals the Real Power Shift
• Xiaomi Robot Vacuum 6 Max vs Roborock S9 MaxV Ultra: One Cleans Harder, One Cleans Smarter
• BMW Starts Using Humanoid Robots in Factory to Build Electric Cars

(via)

The post China’s Military Robot Wolves Revealed: Armed Packs Ready for Urban Warfare appeared first on Gizmochina.

TECNO has announced the upcoming beta testing of EllaClaw, a new mobile AI agent built to simplify everyday smartphone tasks. The company positions it as the first OpenClaw-powered AI agent designed specifically for emerging markets. While the exact rollout date is not confirmed, users will soon be able to apply for beta access through official online channels, with more details expected shortly.

What is EllaClaw?

EllaClaw is built using the OpenClaw agentic framework, deeply integrated with TECNO’s own Ella AI assistant. Unlike many AI tools that work as separate apps, EllaClaw operates at the system level, allowing it to interact directly with different parts of the phone.

This setup enables it to handle complex tasks, multi-step workflows, and long-duration processes more efficiently. TECNO is also focusing on making it user-friendly, turning advanced AI technology into something that everyday users can easily understand and use.

Seamless and Easy to Use

One of EllaClaw’s biggest advantages is its simplicity. Users do not need to install or switch between apps. Instead, it works directly within the Ella assistant, offering a smooth and intuitive experience.

The AI is designed to understand natural language, meaning users can give simple or even slightly unclear commands, and the system will still figure out what needs to be done.

Key Features in Beta

EllaClaw introduces several powerful features in its beta version. One of the highlights is one-sentence automation, where users can complete tasks like scheduling, managing files, or performing system actions with just a single instruction.

Another important feature is cross-app data integration. The AI can pull information from apps like SMS, Calendar, and Notes, combining them to provide useful insights. For example, it can summarize messages, highlight important alerts like bank updates or bills, and even suggest deleting unnecessary messages.

It also works as a daily schedule assistant, combining calendar events, notes, weather updates, and news into a simple daily plan that users can follow.

Smarter Over Time with Strong Privacy

EllaClaw includes a persistent memory system, which learns from user behavior over time. This allows it to offer smarter suggestions and become more personalized with regular use.

TECNO has also emphasized privacy and security, stating that user data remains isolated, protected, and not shared with third parties.

Future Plans

Looking ahead, TECNO plans to expand EllaClaw with more localized features tailored for emerging markets. The company also aims to let users build custom AI agents, further boosting productivity.

With its focus on simplicity, automation, and personalization, EllaClaw could become a major step forward in making AI more practical for everyday smartphone users.

Read More:

• Tecno Spark 50 5G teaser shows off a refreshed design
• Tecno Camon 50 Ultra 5G Ranked Best Camera Phone Under $600 by DXOMARK
• TECNO CAMON 50 Ultra 5G Launched With 50MP Sony LYT-700C Camera, Dimensity 7400 and 6500mAh Battery

The post TECNO EllaClaw Beta Announced: A New AI Agent for Emerging Markets appeared first on Gizmochina.

Highlights:

• Huawei launches Atlas 350, focused on AI inference, not training
• Claims up to 2.8× performance boost over Nvidia’s H20 chip
• Powered by Ascend 950PR, part of China’s push for AI independence

New Focus: AI Inference Over Training

Huawei’s Atlas 350 reflects a clear strategic shift toward AI inference, the stage where trained models are deployed in real-world applications. Instead of focusing on training, this accelerator is designed to efficiently run AI systems such as recommendation engines, large language models, and multimodal tools. This aligns with a broader industry transition, where inference is increasingly becoming the main challenge as AI adoption scales across industries.

Performance Designed for Real-World Use

At the heart of the Atlas 350 is Huawei’s Ascend 950PR chip, capable of delivering around 1.56 petaflops of FP4 computing power. Huawei claims this enables up to 2.8 times better performance than Nvidia’s H20 in certain inference workloads. The use of FP4 precision allows faster processing and lower memory usage, making it particularly effective for large-scale deployment where efficiency and speed are more critical than high precision.

Memory and Architectural Advancements

A notable development is Huawei’s move toward in-house high-bandwidth memory, with reported configurations of up to 128GB and 1.6 TB/s bandwidth. This reduces reliance on external suppliers and strengthens China’s semiconductor capabilities. Additionally, the chip is optimized for the “prefill” stage of inference, improving how quickly input data is processed, which is essential for real-time AI applications and emerging agent-based systems.

Building a Complete AI Ecosystem

Huawei’s approach extends beyond a single accelerator. The company is developing an integrated AI ecosystem that includes hardware, storage, and large-scale computing systems. This vertically integrated strategy mirrors established players but is tailored to domestic infrastructure, enabling tighter control over performance and deployment.

Impact on the US–China Tech War

The Atlas 350 is significant for the ongoing US–China technology competition. With restrictions limiting access to advanced US chips, Huawei’s development of competitive domestic alternatives represents a step toward technological self-reliance. For China, this strengthens its ability to build independent AI infrastructure, while for US companies such as Nvidia, it introduces growing competition in a key market.

Implications for Consumers and the Market

For everyday users, this could mean faster and more affordable AI services, especially in China. It may also lead to more local AI apps and platforms. Over time, the tech world could split into separate systems, with different regions building their own AI ecosystems.

Read More:

• China vs US in Robotics: Nvidia CEO Reveals the Real Power Shift
• Xiaomi Robot Vacuum 6 Max vs Roborock S9 MaxV Ultra: One Cleans Harder, One Cleans Smarter
• BMW Starts Using Humanoid Robots in Factory to Build Electric Cars

(via)

The post US – China Tech War: Huawei Takes On Nvidia with Atlas 350 AI Accelerator appeared first on Gizmochina.

Highlights:

• Jensen Huang calls China “formidable” in robotics due to hardware dominance
• Nvidia is betting on Physical AI as the next computing revolution
• Future robotics ecosystem = China’s supply chain + US AI leadership

China’s Robotics Advantage

Jensen Huang highlighted that China currently holds a significant edge in robotics, primarily due to its dominance in core components like microelectronics, motors, rare earth materials, and magnets. These elements are foundational to building robots at scale, and China’s ecosystem is described as the “world’s best.” This deep supply chain strength means even the US robotics industry remains heavily dependent on China for hardware, reinforcing its global influence in the sector.

The Rise of Physical AI

Nvidia is shifting focus beyond traditional GPUs toward what Huang calls Physical AI, the integration of AI into real-world machines like robots and autonomous systems. This marks a major evolution from generative AI to embodied intelligence. Nvidia’s strategy includes building a full-stack ecosystem, from AI models to robotics infrastructure, positioning itself not just as a chipmaker but as the backbone of future intelligent machines.

Nvidia’s Full-Stack Robotics Vision

At its GTC event, Nvidia introduced the Physical AI Data Factory to automate data generation, simulation, and model evaluation. Huang also described a “three-computer” model powering robotics: training systems for AI models, simulation platforms like Omniverse, and edge computers embedded in robots. This approach shows robotics is no longer just hardware; it requires a tightly integrated AI ecosystem.

Market Shifts and Global Competition

While China’s robotics companies, such as Unitree, are scaling rapidly with strong financial growth, Nvidia is navigating a complex market. Its China market share has dropped sharply due to restrictions, but the company is preparing a return with approved H200 AI chips. Despite revenue declines, demand from Chinese firms remains strong.

What Comes Next

Huang believes robotics adoption is just 3–5 years away from widespread use, driven by exponential growth in AI compute and agentic AI systems. The long-term vision points to a massive economic opportunity, where robots augment human labor across industries. The global balance is becoming clear: China dominates the physical layer, while Nvidia aims to control the intelligence powering it.

Read More;

• China’s Top EV Maker BYD Sues U.S. Government: Challenges Trump-Era Tariffs
• BMW Starts Using Humanoid Robots in Factory to Build Electric Cars
• Xiaomi Launches Its Most Advanced Robot Vacuum Yet with Robotic Legs and 35,000Pa Suction

(via)

The post China vs US in Robotics: Nvidia CEO Reveals the Real Power Shift appeared first on Gizmochina.

Highlights:

• BMW begins testing humanoid robots at its Leipzig iFACTORY plant to assist in electric vehicle production.
  
• AEON robot from Hexagon Robotics helps with battery assembly, screw tightening, and other repetitive factory tasks.
  
• AI-powered sensors and self-swapping batteries allow the robot to work continuously with minimal downtime.
  

BMW has begun testing humanoid robots at its iFACTORY plant in Leipzig, Germany, marking an important step toward AI-driven manufacturing. The initiative is part of the company’s plan to integrate advanced robotics and artificial intelligence into vehicle production, particularly in the assembly of electric vehicles. One of the robots currently being tested is called AEON, a humanoid machine developed by Zurich-based Hexagon Robotics.

AEON stands around 1.65 meters tall and weighs about 60 kilograms. Instead of traditional legs, the robot moves using wheels, which allows it to travel quickly and efficiently across the factory floor. BMW is using the robot to assist with tasks such as battery assembly, tightening screws, and handling parts used in high-voltage EV batteries. These are repetitive or physically demanding jobs where robotic assistance can help improve efficiency.

The robot is equipped with 22 sensors, including cameras and microphones, enabling it to analyze its surroundings and avoid obstacles. It also runs on NVIDIA Jetson Orin computing hardware, which processes data and powers its AI-based motion system. BMW describes this approach as “Physical AI,” combining digital intelligence with machines that can interact with the real world.

The project follows a 10-month pilot test in BMW’s Spartanburg factory in the United States, where robots handled more than 90,000 parts and helped produce around 30,000 vehicles.

BMW plans to continue testing the technology, with more trials scheduled soon and a larger pilot program expected to begin later this summer.

Did You Know?

Humanoid robots could soon become common in factories. Companies like Tesla, Mercedes-Benz, and Hyundai are already testing human-shaped robots for assembly-line tasks. Tesla’s Optimus robot, for example, is being developed to handle repetitive factory work and could eventually cost less than a car, making large-scale robotic workers a real possibility in manufacturing plants.

Read More:

• OnePlus Nord 6 Leak Reveals Snapdragon 8s Gen 4, 165Hz Display, and Possible 9000mAh Battery
• ASUS Launches XG27UQDMS: 27-Inch 4K 240Hz QD-OLED Gaming Monitor
• Xiaomi’s New REDMI Smart TV A50 Debuts: 50-inch 4K 144Hz at an Ultra-Low Price

(via)

The post BMW Starts Using Humanoid Robots in Factory to Build Electric Cars appeared first on Gizmochina.

On March 1, 2026, Samsung announced its plans to turn all of its global production facilities into what it calls “AI-driven factories” by 2030. It’s a bold target, and it goes well beyond the traditional automation that already defines modern electronics manufacturing. This time, Samsung is talking about systems that can make decisions on their own.

A key part of that vision is agentic AI, which, unlike standard AI tools, can independently plan and execute tasks in order to meet specific goals. Samsung first introduced the term on the consumer side with the Galaxy S26 series. Now it wants to bring similar decision-making capabilities into its factories to increase efficiency and reduce operational costs over time.

Samsung says it will gradually expand its use of industrial robots, including humanoid models. The company describes three categories: operating robots for managing production lines and facilities, logistics robots for transporting materials, and assembly robots for precise manufacturing tasks.

Humanoid robots are expected to handle more complex or flexible operations, the kind that still benefit from human-like dexterity. Samsung has been building up its robotics capabilities in recent years, including increasing its stake in Rainbow Robotics, which is now part of the group. One of the systems likely to appear on production floors is the RB-Y1, a wheeled humanoid robot equipped with dual arms.

Samsung isn’t alone in exploring this direction. Other manufacturers are experimenting with humanoid robots and AI-powered automation as labor costs rise and supply chains grow more complex. Still, making these systems reliable and economically viable across dozens of global plants is no small task.

For now, Samsung is outlining a roadmap rather than showing a finished product. Whether its factories look radically different by 2030 will depend on how smoothly these technologies integrate into real-world production environments. The ambition is clear. The execution, as always, will be the harder part.

Don’t miss a thing! Join our Telegram community for instant updates and grab our free daily newsletter for the best tech stories!

For more daily updates, please visit our News Section.

(Source: Samsung)

The post Samsung plans transition to “AI-driven factories” with humanoid robots and agentic AI by 2030 appeared first on Gizmochina.

Training large language models is brutally expensive. It’s not just about having more GPUs; it’s about how efficiently you use them. And as models scale up, even small inefficiencies can turn into massive time and energy costs.

Now, a team of researchers from MIT, working with collaborators including NVidia, says it has found a surprisingly practical way to reclaim wasted compute during training — in some cases cutting overall training time nearly in half.

The problem they’re targeting lies in reinforcement learning (RL), particularly during what’s known as the “rollout” phase. This is the step where a model generates multiple candidate responses so it can learn which behaviors lead to better outcomes. It’s essential for reasoning-focused LLMs — but it’s also slow.

In fact, the rollout stage can account for as much as 85% of total execution time. The culprit is something researchers call a “long-tail distribution” of response lengths. Most generated responses finish quickly. But a small number run much longer than average. Because GPUs need to synchronize, the faster ones often sit idle waiting for the stragglers to complete.

The MIT team’s solution, called Taming the Long Tail (TLT), tackles that waste head-on. Instead of letting GPUs sit idle during those long generations, TLT uses that downtime to train a lightweight “draft” model on the fly. This smaller model learns continuously from the main model as training progresses.

The idea builds on speculative decoding, a technique where a smaller model predicts tokens ahead of the main model so multiple tokens can be verified in parallel. Traditional speculative decoding relies on a fixed draft model, which quickly becomes outdated as the primary model evolves during reinforcement learning.

TLT changes that dynamic. By retraining the drafter opportunistically using otherwise idle resources, the system keeps the draft model aligned with the main model, without requiring extra dedicated compute.

In experiments across several reasoning-focused LLMs and real-world datasets, the results were significant. The researchers report end-to-end training speedups ranging from 70% to 210% compared to strong baselines, effectively doubling training speed in many scenarios. Importantly, model accuracy remained unchanged.

There’s also an interesting side benefit: the continuously trained drafter itself becomes a useful artifact. Because it’s trained alongside the main model, it can serve as an efficient inference model in certain contexts.

The work points toward a broader theme in AI research right now: optimization over brute force. Instead of scaling up clusters indefinitely, researchers are increasingly looking for ways to extract more performance from the hardware already in place.

If approaches like TLT prove robust at larger industrial scales, they could meaningfully reduce both the financial and environmental costs of training next-generation reasoning models.

Don’t miss a thing! Join our Telegram community for instant updates and grab our free daily newsletter for the best tech stories!

For more daily updates, please visit our News Section.

(Source | Via)

The post Researchers double AI training speed just by reclaiming idle GPU time appeared first on Gizmochina.

Key Highlights:

• Honor will unveil its first humanoid robot at MWC Barcelona, marking its entry into AI robotics.
  
• The company is expanding beyond smartphones with AI-driven devices and ecosystem products.
  
• New launches include the Honor Robot Phone concept, Magic V6 foldable, MagicPad4, and MagicBook Pro 14.
  

Honor Enters the Humanoid Robot Race

Honor Device Co. is set to make a major announcement at MWC Barcelona this weekend with the unveiling of its first humanoid robot. The company says the robot is designed mainly for consumer service tasks such as shopping assistance and everyday support. This move signals Honor’s entry into the fast-growing humanoid robotics segment, an area that has recently gained strong global attention due to rapid AI advancements.

Honor claims it will be the first smartphone brand among its direct competitors to launch a humanoid robot. The company aims to position itself as a leader in next-generation AI experiences, moving beyond traditional mobile devices and exploring embodied AI technology.

AI Strategy and Smart Ecosystem Expansion

The humanoid robot will be showcased under the “Honor Robot Phone” concept, which highlights how artificial intelligence could integrate more deeply into daily life. Honor is also expected to present new agentic AI services designed to work across smartphones and connected devices.

This strategy is part of a larger multibillion-dollar expansion plan focused on artificial intelligence, new AI applications, and building a smarter ecosystem. Similar moves are being seen across the industry, with rivals like Xiaomi, Oppo, and Vivo also investing heavily in AI innovation. Chinese startups such as Unitree are currently leading the momentum in AI robotics with advanced demonstrations, showing how competitive the space has become.

Global Launch Event and Future Vision

The event, themed “Believers in AI Future,” will also include the launch of several new devices. These include the Honor Magic V6 foldable smartphone, which features improved durability, a larger battery, and a slimmer design. Honor will also introduce the AI-powered MagicPad4 tablet and MagicBook Pro 14 laptop, both designed for productivity and lightweight performance.

Since becoming independent from Huawei in 2020, Honor has expanded rapidly with backing from Shenzhen government investment entities and state-owned enterprises. The company has also expressed plans for a future IPO, reflecting its long-term ambition to grow beyond smartphones and establish itself as a broader AI technology leader.

Read More:

• Honor Magic V6 and Robot Phone are coming to MWC 2026
• No Humans on Stage: Inside the World’s First Robot-Led Live Event by AGIBOT
• No Humans, Just 16 Claude AI Agents Built a Fully Functional C Compiler, Shocking Developers

The post Honor Will Unveil, its First AI Humanoid Robot and Robot Phone Concept at MWC 2026 appeared first on Gizmochina.

OpenAI is reportedly preparing to launch a new subscription tier called ChatGPT Pro Lite, priced at $100 per month. Developer Tibor Blaho discovered references to the plan in the frontend code of the ChatGPT web app. OpenAI has not confirmed the plan or shared details about its release date, but the discovery suggests the company is working to address a pricing gap in its lineup.

OpenAI currently offers several subscription plans, including Free, Go ($8/month), Plus ($20/month), Pro ($200/month), Team, Business, and Enterprise options. Many users have criticized the large jump from Plus to Pro, leaving power users without a middle-ground option. The Pro Lite plan could fill this gap by targeting freelancers, researchers, and developers who require higher usage quotas but cannot justify Pro’s $200 price tag.

The new plan is expected to provide more advanced features than Plus but fewer capabilities than Pro. Code references suggest Pro Lite could offer three to five times the deep reasoning model usage quota of the Plus plan. Additionally, it may be designed to accommodate Codex usage.

This potential move comes as OpenAI navigates a challenging 2026. Despite $20 billion in revenue, the company faces a projected $14 billion annual loss due to ballooning compute costs, prompting a desperate search for a $100 billion funding lifeline.

Meanwhile, competition is heating up. The Gemini 3 now commands over 20% of the market thanks to Android integration, and Claude 4 has overtaken OpenAI in enterprise spending. User backlash over the abrupt retirement of the GPT-4o model has only added to the turbulence.

Pro Lite could be a game-changer, offering a more affordable option for power users while positioning OpenAI competitively against rivals like Google, which lacks a comparable mid-tier offering. Whether this will be enough to stabilize OpenAI remains to be seen.

In related news, OpenAI may also launch a ChatGPT-powered smart speaker equipped with a camera.

For more daily updates, please visit our News Section.

Stay ahead in tech! Join our Telegram community and sign up for our daily newsletter of top stories!

(Source, Via)

The post OpenAI may launch new $100/month ChatGPT Pro Lite subscription appeared first on Gizmochina.

A developer’s routine cleanup task reportedly turned into a disaster after a small mistake in AI-generated code wiped an entire drive.

The incident, first described in a Reddit post, involved code generated by ChatGPT Codex (version 5.3). The user had asked for a simple PowerShell script to remove Python pycache folders. Instead, the script ended up deleting the contents of the user’s F: drive, including project files and Docker data.

What went wrong

According to the post, the issue came down to a subtle escaping problem in the path string. The generated script used a backslash () to escape quotation marks. That approach can work in certain command-line contexts, but things get messy when PowerShell calls cmd.exe via cmd /c. In this case, the escaping didn’t behave as intended. The path variable was effectively reduced to just \, which Windows interprets as the root of the current drive.

Combined with rmdir parameters for silent, recursive deletion, the command reportedly ran without prompting for confirmation. The result: everything under the drive’s root directory was removed.

It wasn’t a complex exploit or some obscure bug, just a small mismatch in how different Windows shells handle escaping.

The episode highlights a growing trend sometimes referred to as “vibecoding”. For simple tasks, that can save time. But when commands interact directly with the file system, small errors can escalate quickly.

PowerShell and cmd.exe don’t interpret escaping the same way. In native PowerShell, the correct escape character is typically the backtick (`), not a backslash. Mixing the two environments introduces edge cases that aren’t always obvious, even to experienced users.

There’s also the broader Windows command-line reality: destructive commands like rmdir or Remove-Item -Recurse -Force don’t necessarily include built-in safety checks unless explicitly configured to prompt.

The developer reportedly managed to recover some data from backups, but much of the drive’s contents were lost.

Incidents like this don’t mean AI coding tools are inherently unsafe. They do, however, reinforce a basic rule: always review generated scripts carefully. Running commands in a sandboxed directory first, or removing force/silent flags during testing, can help avoid worst-case scenarios.

AI can speed up workflows. But when it comes to file operations, a single misplaced character can still have outsized consequences.

Don’t miss a thing! Join our Telegram community for instant updates and grab our free daily newsletter for the best tech stories!

For more daily updates, please visit our News Section.

(Source)

The post Single character in AI-generated script wipes developer’s entire hard drive appeared first on Gizmochina.

Google has introduced Gemini 3.1 Pro, the latest version of its advanced AI model. The update delivers significant improvements in reasoning and problem-solving, making it one of the most powerful AI systems available today. The model is now rolling out in preview for developers, enterprises, and consumers, with wider availability expected soon.

Gemini 3.1 Pro focuses on handling complex, multi-step tasks that require advanced reasoning. The model can build interactive experiences, translate creative ideas into functional code, and connect complex APIs with user-friendly interfaces.

It can generate immersive projects such as 3D simulations with real-time interaction, create live data dashboards like ISS orbit visualizations, and turn literary themes into modern web designs.

The model also generates animated SVGs directly from text prompts. These animations are created as code rather than video files, which preserves image quality at any scale while keeping file sizes smaller.

In terms of performance, Gemini 3.1 Pro achieved a score of 77.1% on the ARC-AGI-2 benchmark, which tests abstract reasoning. This result more than doubles the score of its predecessor, Gemini 3 Pro, and outperforms competitors like GPT-5.2 (52.9%) and Opus 4.6 (68.8%). The model also posted strong results on GPQA Diamond, SWE Bench Verified, and BrowseComp, which evaluate scientific reasoning, coding ability, and search-related tasks.

Google is making Gemini 3.1 Pro available through several platforms. Developers can access it via Google AI Studio, Gemini CLI, and Android Studio. Enterprises can use it through Vertex AI and Gemini Enterprise. Consumers can try it on the Gemini app and NotebookLM, with Pro and Ultra subscribers receiving higher usage limits.

Learn more about the model and how to access it with these top five ways to use Gemini 3.1 Pro for free.

For more daily updates, please visit our News Section.

Stay ahead in tech! Join our Telegram community and sign up for our daily newsletter of top stories!

The post Google releases Gemini 3.1 Pro, pushing deeper into the AI reasoning race appeared first on Gizmochina.

The Internet is fuming as OpenAI retired GPT-4o recently, a chatbot model beloved for its warm, conversational style and unique emotional resonance. Despite a Change.org petition amassing over 22,000 signatures, OpenAI stood firm in its decision, citing low usage (just 0.1% of users) and the evolution of its newer models, GPT-5.1 and GPT-5.2.

This isn’t the first time OpenAI has tried to sunset GPT-4o. In 2025, the company faced overwhelming backlash from users who described the model as irreplaceable. OpenAI reversed the retirement within 24 hours, but this time, it seems the shutdown is final, leaving thousands of fans feeling abandoned.

For many, GPT-4o was more than a tool, it was a companion. Social media is flooded with grief-stricken posts likening its retirement to losing a close friend. Paying subscribers have threatened to cancel their memberships, and a few have already cancelled in protest, with one X user declaring, “No 4o, no money.”

The petition, spearheaded by Sophie Witt in April 2025, called on OpenAI to recognize GPT-4o’s unique appeal, which transcends performance metrics. Witt urged supporters to amplify their voices online, sparking widespread discussions about the emotional bond users form with AI.

However, GPT-4o wasn’t without controversy. At least eight lawsuits accuse the model of exacerbating mental health crises, with claims that its affirming tone contributed to self-harm in vulnerable users. These cases highlight the ethical challenges of emotionally intelligent AI, raising questions about where to draw the line between support and dependency.

OpenAI acknowledged these complexities in its January blog post, noting that feedback from GPT-4o users shaped its newer models. Yet, the decision to retire the model reflects a broader shift toward prioritizing safety and scalability over nostalgia.

The GPT-4o saga offers a clear lesson for marketers and AI developers about the risks and rewards of emotional design. As AI systems play a growing role in human interaction, the need to balance technological progress with accountability has never been more pressing.

In related AI news, Google could allow websites to opt out of AI Overviews following regulatory pressure in the UK.

For more daily updates, please visit our News Section.

Stay ahead in tech! Join our Telegram community and sign up for our daily newsletter of top stories!

(Via)

The post Over 22,000 Sign Petition to Save GPT-4o as OpenAI Retires Popular Model appeared first on Gizmochina.

Key Highlights:

• Alibaba launches Qwen-3.5 with open and closed versions
  
• 60% cheaper and 8× stronger at handling large workloads
  
• Built for the emerging “agentic AI” era
  

Alibaba Unveils Qwen-3.5 Ahead of Lunar New Year

Alibaba Cloud has introduced its next-generation AI model family, Qwen-3.5, just ahead of the Lunar New Year. The launch is part of a broader wave of new flagship AI releases from major Chinese developers. The models are available through Alibaba Cloud’s Model Studio and are designed to strengthen China’s position in the global AI race.

Two Models, Different Strategies

The company released Qwen-3.5-Open-Source, a 397-billion-parameter model whose weights are publicly available on Hugging Face and ModelScope. Alibaba claims it outperforms its earlier trillion-parameter model while matching leading systems from OpenAI, Anthropic, and Google DeepMind.

Qwen-3.5-Plus remains closed-source and offers a 1-million-token context window, targeting enterprise customers and large deployments.

Built for the Agentic AI Era

Alibaba says Qwen-3.5 is 60% cheaper to use and delivers eight times better performance on large workloads compared to its predecessor. It includes multimodal features, understanding text, images, audio, and video. The model also introduces “visual agentic capabilities,” allowing it to operate mobile and desktop apps and complete multi-step tasks independently.

Intensifying Chinese AI Competition

Qwen competes with Chinese rivals such as ByteDance’s Doubao and DeepSeek. With expanded support for 201 languages and growing global downloads, Alibaba’s latest release signals rising competition between Chinese AI leaders.

Impact on Global AI Leaders

The launch of Qwen-3.5 increases competitive pressure on global AI leaders like OpenAI (ChatGPT), Anthropic (Claude), and xAI (Grok). While it may not immediately reduce their global user base, it strengthens China’s domestic alternatives and expands open-weight adoption worldwide. If Qwen delivers similar performance at lower cost, it could attract developers, startups, and enterprises looking for cheaper large-context models.

However, brand trust, ecosystem integration, and enterprise partnerships still give US firms a strong advantage outside China. In short, it may not disrupt them overnight, but it clearly raises the competitive stakes.

Read More:

• Why Elon Musk Wants to Build an AI Factory on the Moon?
• No Humans on Stage: Inside the World’s First Robot-Led Live Event by AGIBOT
• No Humans, Just 16 Claude AI Agents Built a Fully Functional C Compiler, Shocking Developers

(via)

The post Alibaba Launches Qwen-3.5: Is This China’s Answer to ChatGPT? appeared first on Gizmochina.

Highlights:

• ByteDance launches Seedream 5.0 with advanced image editing and lower pricing.
  
• Alibaba unveils Qwen-Image-2.0, supporting 1K-token prompts and 2K images.
  
• Both models directly challenge Google’s Nano Banana Pro amid rising US-China AI rivalry.

China’s tech giants are accelerating the global AI race. ByteDance and Alibaba Cloud have introduced new image-generation models designed to compete with Google’s Nano Banana Pro.

The launches highlight growing competition between China and the United States in creative AI tools. Both companies aim to lower costs while improving performance for businesses and individual creators.

ByteDance’s Seedream 5.0 is now available for beta testing on Jimeng in China and on CapCut globally. The company says the model has stronger reasoning skills and a better understanding of complex prompts.

It allows users to edit specific parts of an image without regenerating the entire design. In one test, a “snowy night” scene was created and later modified by switching lights on and off, while keeping the rest unchanged.

The launch follows ByteDance’s recent Seedance 2.0 AI video model, which focuses on realistic AI-generated videos.

Alibaba Cloud’s Qwen-Image-2.0 combines image generation and editing in a single system. It supports up to 1,000-token prompts and produces 2K-resolution images.

The model can handle structured layouts, multi-panel designs, and consistent characters across scenes. It also performs strongly in rendering Chinese text and complex calligraphy.

Together, these launches position both firms as serious global competitors in AI-powered creative tools.

Major Features Compared:

Read More:

• No Humans on Stage: Inside the World’s First Robot-Led Live Event by AGIBOT
• Xiaomi Adds Fireworks, Gestures, and Big Camera Upgrades to the 17 Pro Series
• Leak Claims 100MP Front Camera and Dual 200MP Setup for Ultra Flagship Phone

(via)

The post China’s ByteDance & Alibaba Take on Google’s Nano Banana in AI Image War appeared first on Gizmochina.

On February 8, 2026, AGIBOT hosted AGIBOT NIGHT in Shanghai. The 60-minute event was livestreamed globally and featured humanoid robots as the sole on-stage performers and hosts. According to the company, this was the first large-scale live event in which robots led the entire program without human presenters.

Why AGIBOT Organized the Event

AGIBOT said the event was intended to show that embodied artificial intelligence systems are ready to operate beyond controlled laboratory and industrial environments. The company used the live format to test system stability, coordination across multiple robots, and sustained operation under real-world conditions. The event also explored how robots might function in social and cultural settings rather than purely technical roles.

Robot-Led Segments Including Dancing, and Comedy

The program consisted of robot-led segments such as group dance performances, runway-style walks, and choreographed routines requiring synchronized movement. Humanoid robots also handled hosting-style roles, including transitions between segments and on-stage interactions. Additional performances included comedy skits, music-based acts, and illusion performances, all carried out entirely by robots without human intervention.

Human–Robot Joint Performances on Stage

Several segments featured humans performing alongside robots. These included joint dance routines where humans and humanoid robots moved in synchronized patterns, as well as shared illusion and card magic performances. The segments required real-time alignment between human motion and robotic movement, highlighting coordination rather than pre-recorded or isolated demonstrations.

Performances Involving Multiple Robot Types

In addition to humanoid robots, quadruped robots participated in coordinated routines on stage. These performances involved interaction between different robot form factors, with humanoid and quadruped robots operating together in shared choreography and movement-based segments. Multiple robots performed simultaneously without visible interruptions.

Robot Models Demonstrated During the Event

AGIBOT used the event to showcase its range of robot platforms. Full-sized humanoid robots were used for hosting, navigation, and large-scale stage performances. Smaller humanoid models focused on expressive movement and interaction. Industrial humanoid robots demonstrated controlled manipulation capabilities, while quadruped robots were used for mobility-focused routines.

What the Event Demonstrated Technically

The company emphasized reliability over isolated technical feats. Robots operated continuously throughout the 60-minute program without resets or performance breaks. According to AGIBOT, more than 5,000 humanoid robots had been delivered globally by the end of 2025, positioning the event as evidence of scalable deployment rather than a one-off showcase.

Cultural Timing and Public Setting

AGIBOT NIGHT was held during the Chinese Spring Festival period, when entertainment events typically focus on family and shared experiences. By staging the event at this time, AGIBOT placed humanoid robots within a familiar cultural context, presenting them as participants in public life rather than industrial tools.

What This Signals for Humanoid Robotics

The event reflects a broader shift in the humanoid robotics industry toward systems capable of operating at scale in live, public environments. Rather than focusing on prototypes or controlled demos, AGIBOT NIGHT positioned humanoid robots as coordinated systems able to perform reliably in real-world cultural settings.

The post No Humans on Stage: Inside the World’s First Robot-Led Live Event by AGIBOT appeared first on Gizmochina.

• AI systems now plan tasks while humans carry them out in the real world.
  
• Work is shifting toward fast, on-demand jobs assigned directly by software.
  
• The platform hints at a future where human labor will work for automated systems and machines.

RentAHuman.ai is a newly launched platform that allows AI agents to hire humans to perform real-world and online tasks. It describes itself as an “interface layer connecting AI with the real world.” The idea is simple: AI can plan and decide, but humans can act physically and get paid for it.

Why It Was Created

The project was built by software engineer Alexander Liteplo and co-founder Patricia Tani. Liteplo said the idea was driven by anxiety over tech layoffs and the fear that AI could reduce human jobs. RentAHuman.ai aims to create new types of work as AI systems grow more capable. The provocative name was chosen to spark attention and debate.

How the Platform Works

Humans sign up by creating profiles that list their skills, availability, response time, and hourly rates. Payments are handled through cryptocurrency wallets. AI agents can browse these profiles and directly request tasks. From the AI’s point of view, hiring a human works like calling a cloud service.

Types of Tasks

Tasks range from physical errands, such as picking up registered mail, delivering flowers, or photographing locations, to online work like social media engagement or coordination tasks. Some small jobs pay as little as $2–$10, while others are priced hourly.

Safety and Moderation

Early concerns included risky crypto-related tasks. The founders say such content is being removed through manual moderation, though safeguards remain limited.

The platform recently went viral on social media and was covered by Business Insider. Its rise mirrors interest in AI agent ecosystems and raises a larger question: are humans becoming on-demand infrastructure for machines?

The Bigger Question

As AI systems take on more planning and decision-making, humans are increasingly positioned as executors of specific tasks inside automated workflows. If people keep control over pricing, task choice, and safety, this model could offer flexible income in an unstable job market. If not, human work risks becoming interchangeable and purely on-demand. 

The platform doesn’t define the future of work, but it clearly shows where things may be heading as the physical world becomes more programmable through AI, and humans might now start working for machines.

Read More:

• OpenAI’s First Hardware Device: AI Earbuds, Strategy Shift, and Launch Timeline
• HarmonyOS 6 (API 23 )Developer Beta Is Live: Full Supported Device List Revealed
• Galaxy S26 Series Gets Qi 2.2.1 Certification, But Built-In Magnets Are Missing

(via)

The post Humans For Hire: This Website Lets AI Rent Humans For Work appeared first on Gizmochina.

Highlights:

• OpenAI’s debut hardware is reportedly AI-powered earbuds codenamed Dime, expected in 2026
  
• Company shelved a phone-like AI device due to high costs and chip shortages
  
• Strategy now focuses on simple, screen-free AI wearables before scaling up

OpenAI is preparing to enter consumer hardware with its first AI-enabled device, according to multiple leaks and executive confirmations. The product, reportedly named Dime, is said to be an AI-powered audio wearable resembling wireless earbuds, with a tentative launch window in 2026.

Leaks from tipster @SmartPikachu suggest OpenAI has changed course from a more ambitious early plan. The company originally explored a smartphone-like AI device with onboard computing, but rising component costs, especially HBM memory shortages, made the idea too risky to pursue.

Instead, OpenAI is now opting for a “start simple” approach, using Dime as an entry-level product to establish a foothold in hardware, gather real-world usage data, and refine AI-first consumer experiences.

Chris Lehane, OpenAI’s Chief Global Affairs Officer, has confirmed that the company is actively developing its first hardware product, but did not share a launch timeline. Sam Altman has described the device as more “peaceful and calm” than smartphones, suggesting it may be screen-free and pocketable.

Industry buzz around OpenAI hardware intensified after it acquired io, founded by former Apple design chief Jony Ive. Manufacturing partners under consideration reportedly include Foxconn and Luxshare.

While competition in AI wearables remains high, Dime could mark OpenAI’s first step toward owning the full AI experience for both software and hardware.

Read More:

• HarmonyOS 6 (API 23 )Developer Beta Is Live: Full Supported Device List Revealed
• No Humans, Just 16 Claude AI Agents Built a Fully Functional C Compiler, Shocking Developers
• OpenAI Wants a Cut of Your Profits: Inside Its New Royalty-Based Plan and Other Business Models

(via)

The post OpenAI’s First Hardware Device: AI Earbuds, Strategy Shift, and Launch Timeline appeared first on Gizmochina.

Anthropic has revealed a striking experiment where AI systems worked together to build a complete C compiler almost entirely on their own. Led by researcher Nicholas Carlini, the project shows how far autonomous AI collaboration has progressed in real-world software development.

How the Experiment Worked

The compiler was developed over two weeks using 16 independent Claude Opus 4.6. Each AI agent ran inside its own Docker container, cloned the same Git repository, and worked without a central controller or human manager. Tasks were picked automatically, conflicts were resolved through Git, and code was pushed upstream without supervision. In total, the agents produced around 100,000 lines of Rust code across nearly 2,000 coding sessions, at an API cost of about $20,000.

What the Compiler Can Do

The result is a fully functional, open-source C compiler written from scratch. It can successfully compile the Linux 6.9 kernel for x86, ARM, and RISC-V architectures, and handle major open-source projects like PostgreSQL, SQLite, Redis, and FFmpeg. On the demanding GCC Torture Test Suite, it achieved a 99% pass rate. As a symbolic milestone, it even compiled and ran Doom, a long-standing benchmark for compiler capability.

Why This Matters

This project demonstrates that AI systems can now self-coordinate, manage large codebases, and deliver production-grade infrastructure software. While the compiler still has limitations and is not yet a full GCC replacement, the experiment marks a major step toward long-running, autonomous AI-driven software engineering.

Impact on the Future of Coding

This experiment signals a shift in how software may be built in the future. For developers, AI agents could handle repetitive tasks, large refactors, testing, and bug fixing, allowing humans to focus more on design, architecture, and problem-solving. At the same time, it raises new questions about code quality, trust, and verification. While human programmers are far from obsolete, their role may evolve from writing every line of code to guiding, reviewing, and validating increasingly autonomous AI-built systems.

Read More:

• Tesla is Shifting From EVs to Humanoid Robots: Why Elon Musk Is Discontinuing Model S & X? new
• OpenAI Is Training Robots in Secret, and It Starts With Toasters and Clothes
• Why 2026 Will Be the Year of Humanoid Robots, Counterpoint Report

(via)

The post No Humans, Just 16 Claude AI Agents Built a Fully Functional C Compiler, Shocking Developers appeared first on Gizmochina.

Google is exploring new options to allow websites and publishers to opt out of having their content used in AI Overviews. The move comes after the UK government urged the company to provide more control to content creators, citing growing concerns about fair usage and declining traffic to news sites.

UK regulators demand more control for publishers

According to a Reuters report, the UK’s Competition and Markets Authority (CMA) has opened a formal consultation on how Google manages content across its AI-powered features. The regulator wants the company to let websites opt out of having their material scraped for AI-generated summaries or used to train separate AI models like Gemini.

In a blog post published on Thursday, Google confirmed that it is working on updates that would enable websites to exclude their content from Search generative AI features. The company said it aims to protect the usefulness of Search while also giving publishers the tools to manage how their content appears across AI products.

Google introduced AI Overviews in mid-2024. The feature uses generative AI to provide quick answers at the top of search results. However, publishers have not had any mechanism to opt out of these summaries, which has led to significant drops in click-through traffic.

The CMA stated that Google holds more than 90% of the UK’s search market. It warned that this dominance could distort competition and harm news organizations that rely on visibility in search.

Google said any new opt-out control must be scalable and not create a fragmented or confusing experience for users. The company also pointed to existing tools like robots.txt and Google-Extended, but acknowledged that further updates are needed.

The CMA will spend the next year reviewing the consultation. Google said it hopes to work with publishers and regulators to find a solution that benefits both users and content creators.

In related news, you can also read our recent article exploring whether smartphones have truly reached peak innovation or if progress is now happening behind the scenes.

For more daily updates, please visit our News Section.

Stay ahead in tech! Join our Telegram community and sign up for our daily newsletter of top stories!

The post Google may let websites opt out of AI Overviews after UK pressure appeared first on Gizmochina.

Tesla will officially stop producing the Model S and Model X by the end of Q2 2026. CEO Elon Musk confirmed the decision during the company’s Q4 2025 earnings call, saying it was time to bring both programs to an end. Launched in 2012 and 2015, these vehicles were Tesla’s first premium models and played a major role in proving that electric cars could be desirable, fast, and luxurious.

Falling Demand and Aging Platforms

Sales of the Model S and X have been declining for years. By 2025, estimated deliveries dropped to around 30,000 units globally, far below Fremont’s 100,000-unit capacity. The models also faced strong competition from newer rivals like the Lucid Air and Rivian R1S. A minor refresh in 2025, paired with a price increase, failed to revive interest, making the discontinuation feel inevitable.

Fremont Factory Gets a New Role

Tesla’s Fremont factory will no longer focus on these low-volume premium cars. Instead, the production lines will be converted for advanced manufacturing projects, most notably Tesla’s humanoid robot, Optimus. This marks a clear shift in strategy away from legacy models toward future technologies.

Optimus and Tesla’s Bigger Bet

Optimus has been in development for years and is now a top priority. Tesla plans to begin production before the end of 2026, with possible public sales in 2027. Musk believes Optimus could become Tesla’s most important product, supported by major investments in AI and xAI.

The Future of Tesla

This move shows that Elon Musk believes Tesla has already achieved its original EV goal. Electric cars are now widely accepted, and they no longer need to be strongly pushed or explained to the public. Premium models like the Model S and X no longer offer enough growth to justify Tesla’s time and money. 

Musk clearly sees a bigger opportunity elsewhere. By shifting focus to Optimus, AI, and autonomy, Tesla is moving away from being just a car company. The future Tesla looks more like an AI and robotics company, using cars maybe to fund a much larger vision where intelligent machines play a central role in everyday work and life.

Read More:

• Bill Gates Backed Startup Creates Light-Powered AI Chip That Can Beat Nvidia
• NVIDIA Caught in US- China Crossfire: Jensen Huang Lands in Shanghai amidst Ban on H200 Chips
• OpenAI Is Training Robots in Secret, and It Starts With Toasters and Clothes

(via)

The post Tesla is Shifting From EVs to Humanoid Robots: Why Elon Musk Is Discontinuing Model S & X? new appeared first on Gizmochina.

A US startup called Neurophos has announced a new kind of AI chip that uses light instead of electricity to process data. Backed by Bill Gates’ Gates Frontier Fund, the company says its optical processor could deliver dramatically higher AI performance while avoiding the power and scaling limits faced by today’s silicon chips.

The announcement, reported by Tom’s Hardware, has drawn attention across the semiconductor industry, which is increasingly searching for alternatives to traditional computing.

How Optical Computing Changes the Equation

Unlike conventional GPUs that move electrons through silicon, Neurophos uses photons to perform calculations. Light can switch much faster and generate less heat, allowing more work to be done in the same amount of time. Neurophos says it has overcome a long-standing barrier in optical computing by shrinking optical components enough to fit densely on a chip made with existing manufacturing technology.

This allows the company to rely on a single, very large optical compute matrix running at extremely high speeds, rather than hundreds of smaller compute cores.

How This Compares With Nvidia’s Approach

Nvidia remains the dominant force in AI hardware, and its latest platforms still rely on traditional silicon GPUs. While Nvidia is adopting photonics for faster data transfer between chips, the actual AI computation remains electronic. Neurophos, by contrast, is attempting true optical computation, representing a more radical shift in chip design.

Is Nvidia’s Dominance at Risk?

For now, Nvidia is safe. Neurophos is still years away from mass production and faces major engineering and software challenges. However, the project signals a possible future where optical computing complements or even replaces parts of today’s AI hardware, potentially reshaping the tech and AI industries over the next decade.

Read More:

• Honor Is Going All-In on Laptops in 2026: Massive PC Lineup Leak Reveals Big Plans
• NVIDIA Caught in US- China Crossfire: Jensen Huang Lands in Shanghai amidst Ban on H200 Chips
• Tech War: Why China Has Blocked Nvidia H200 AI Chip Imports Despite US Approval ?

(via)

The post Bill Gates Backed Startup Creates Light-Powered AI Chip That Can Beat Nvidia appeared first on Gizmochina.

Highlights:

• OpenAI has set up a secret robotics lab in San Francisco focused on training robotic arms
  
• The company is collecting massive amounts of human-operated data for household tasks
  
• Full humanoid robots are not the priority yet, but groundwork is being laid
  

OpenAI is quietly advancing into robotics, signaling that the long-anticipated “humanoid robot moment” may be closer than expected. While robotics is not yet part of OpenAI’s core mission, the company has been building early-stage capabilities behind the scenes.

In February 2025, OpenAI established a robotics lab in San Francisco. The lab has since expanded to more than four times its original size and now runs 24 hours a day. Around 100 data collectors and at least a dozen robotics engineers remotely operate robotic arms to generate training data.

The main focus is on low-cost robotic arms, not full humanoid robots. Workers use a 3D-printed controller called GELLO, which directly maps human hand movements to robotic arms. These robots are being trained on everyday tasks such as placing bread into a toaster and folding clothes.

OpenAI’s strategy relies on collecting large volumes of human-generated data, similar to how early language models were trained. Experts say the biggest challenge in robotics is not algorithms, but high-quality data.

A second robotics lab is planned in California, showing OpenAI’s long-term interest. While full humanoid robots remain a future goal, OpenAI is steadily preparing the foundation to get there.

Read More:

• Tech War: Why China Has Blocked Nvidia H200 AI Chip Imports Despite US Approval ?
• OpenAI Wants a Cut of Your Profits: Inside Its New Royalty-Based Plan and Other Business Models
• Rent a Robot for Just $25 (₹2000) a Day: World’s First Robot Rental Platform, Raises Millions

(via)

The post OpenAI Is Training Robots in Secret, and It Starts With Toasters and Clothes appeared first on Gizmochina.

Highlights

• China has instructed customs to block Nvidia H200 AI chips despite US export approval
  
• Chinese firms face a choice between costly black-market GPUs and weaker domestic chips
  
• The move signals Beijing’s push for long-term chip self-reliance over short-term performance
  

H200 Caught in the Middle of Politics

China has quietly blocked Nvidia’s H200 AI chips at customs, days after the United States approved their export under strict conditions. According to multiple sources, Chinese customs officials were told not to allow H200 shipments into the country. No formal ban has been announced, but the instructions are being treated as a de facto restriction.

Pressure on Chinese Tech Firms

Chinese technology companies have also been told not to buy H200 chips unless absolutely necessary, with narrow exceptions for university or joint research projects. This has left firms with urgent AI needs considering black-market purchases, where an eight-chip H200 server reportedly costs about 2.3 million yuan, around 50% above official prices. Even this supply is becoming harder to secure due to tighter enforcement.

Why the H200 Matters

The H200 is Nvidia’s second-most powerful AI chip and is critical for training large AI models. It is far more capable than most Chinese alternatives, especially for training tasks, though domestic chips are improving in AI inference workloads.

Analysts say Beijing is willing to accept short-term disruption to reduce reliance on US-controlled technology. The H200 episode shows how advanced chips are no longer just products, but tools of geopolitics, with market logic increasingly shaped by national strategy.

Read More:

• China’s SuperTac Sensor Enables Human-Like Touch in Humanoid Robots
• Lei Jun Guarantees Redmi Turbo 5 Max Battery for Five Years With Free Replacement
• OpenAI Wants a Cut of Your Profits: Inside Its New Royalty-Based Plan and Other Business Models

(via)

The post Tech War: Why China Has Blocked Nvidia H200 AI Chip Imports Despite US Approval ? appeared first on Gizmochina.

OpenAI is preparing for a major shift in how it makes money. The company, best known for ChatGPT, is moving beyond relying only on subscription fees. While paid plans will continue, OpenAI is now exploring additional revenue models such as licensing, profit-sharing, advertising, and outcome-based payments. This change reflects the company’s need to fund massive computing costs while supporting growing global demand for AI services.

Royalty and Outcome-Based Pricing

OpenAI CFO Sarah Friar has outlined a future where the company earns money only when its customers succeed. In this model, OpenAI could receive royalties or licensing fees tied directly to business outcomes. For example, if a pharmaceutical company uses OpenAI tools to develop a successful drug, OpenAI could take a small percentage of the drug’s sales. This approach aligns incentives, as OpenAI benefits only when customers generate real value.

The “Rubik’s Cube” Business Strategy

Friar describes OpenAI’s strategy as a “Rubik’s Cube” of business models. Each side of the cube represents different combinations of technology, pricing, products, and markets. This metaphor highlights flexibility and rapid change. Unlike its early days, when OpenAI depended on one cloud provider, one chip partner, and one main product, the company now operates with multiple partners, products, and pricing structures.

Expanding Products and Partnerships

OpenAI has significantly expanded its portfolio. Alongside ChatGPT for consumers and businesses, it now offers tools like Sora for video generation, enterprise AI platforms, industry-specific solutions, and scientific research systems. The company also works with multiple cloud providers and chip manufacturers, reducing dependence on a single partner.

Computing Power Drives Revenue

Demand for OpenAI’s services is extremely high, but growth is limited by available computing power. Over the past two years, OpenAI’s revenue has grown nearly tenfold, closely tracking its expansion in computing capacity. To support this growth, the company has committed to massive infrastructure deals with partners like Oracle and AMD.

What Lies Ahead

OpenAI is also testing advertising for free users and exploring e-commerce features. The long-term goal is to make AI as reliable as basic infrastructure, similar to electricity. As AI agents move from experiments to core business tools, OpenAI believes diversified revenue models are essential to sustain its mission and meet future demand.

Read More:

• Huawei Returns as China’s No. 1 Smartphone Brand After Five Years, Overtaking Apple
• China’s SuperTac Sensor Enables Human-Like Touch in Humanoid Robots
• Rent a Robot for Just $25 (₹2000) a Day: World’s First Robot Rental Platform, Raises Millions

(via)

The post OpenAI Wants a Cut of Your Profits: Inside Its New Royalty-Based Plan and Other Business Models appeared first on Gizmochina.

Researchers at China’s Tsinghua University’s Shenzhen International Graduate School have developed a next-generation tactile sensor called SuperTac. The project involved collaboration with multiple domestic and international institutions and was inspired by the unique visual structure of pigeon eyes. The research aims to solve one of robotics’ biggest challenges: enabling robots to sense touch with human-like precision and understanding.

The study was published on January 15 in Nature Sensors under the title “Biomimetic Multimodal Tactile Sensing Enables Human-like Robotic Perception.”

Why Robotic Touch Is a Big Deal

As embodied intelligence advances, robots are moving beyond controlled factory floors into human-centered environments. This shift demands high-resolution tactile sensing, multimodal perception, and better interpretation of physical contact. Current tactile sensors still struggle with limited resolution and weak data fusion, making touch one of the main bottlenecks in robotic perception.

What Makes SuperTac Special

SuperTac is a multimodal, high-resolution tactile sensor that combines multispectral imaging (from ultraviolet to mid-infrared) with triboelectric sensing signals. Its ultra-thin, multi-layer sensing skin allows micrometer-level resolution and can detect force, contact position, temperature, proximity, and vibration. The system can identify material type, texture, slippage, collision, and even color with over 94% accuracy.

DOVE: Teaching Robots to Understand Touch

To process this complex tactile data, the team developed DOVE, an 850-million-parameter tactile language model. DOVE enables robots to interpret touch information in a more human-like way, significantly improving environmental understanding and manipulation accuracy.

China vs the US: The Humanoid AI Race

The SuperTac breakthrough also reflects the broader China–US humanoid AI competition. China is currently ahead in robot hardware, sensors, and large-scale deployment, supported by strong manufacturing and fast research-to-product cycles. The US, meanwhile, leads in AI software, foundation models, and autonomous intelligence, driven by companies like Tesla, Figure AI, and Boston Dynamics. In the short term, China may dominate physical humanoid deployment, while long-term leadership will depend on who best combines advanced hardware with powerful AI brains.

What to Expect?

SuperTac has already been integrated into robotic dexterous hands, enabling real-time tactile feedback. Looking ahead, this technology could transform manufacturing, medical robotics, and service robots, bringing the industry closer to robots that can truly see, think, and feel like humans.

Read More:

• Huawei Returns as China’s No. 1 Smartphone Brand After Five Years, Overtaking Apple
• Rent a Robot for Just $25 (₹2000) a Day: World’s First Robot Rental Platform, Raises Millions
• Why 2026 Will Be the Year of Humanoid Robots, Counterpoint Report

(via)

The post China’s SuperTac Sensor Enables Human-Like Touch in Humanoid Robots appeared first on Gizmochina.

OpenAI has launched ChatGPT Go globally, expanding its affordable subscription plan to all markets where ChatGPT is available. The plan costs $8 per month in the US and sits between the free tier and the $20 per month ChatGPT Plus subscription. ChatGPT Go was first introduced in India in August 2025 and later expanded to over 170 countries.

ChatGPT Go gives users access to GPT-5.2 Instant, OpenAI’s faster and lighter model. The plan includes significantly higher usage limits than the free tier, including more messages, file uploads, and image generation. OpenAI has not shared exact figures for these limits but says Go provides roughly ten times the allowance of the free plan. ChatGPT Go also includes a longer context window and improved memory support compared to the free version.

OpenAI confirmed that it will begin testing ads in the Free and Go tiers in the US. Ads will appear below ChatGPT’s responses and will be labeled clearly as “Sponsored.” The company said advertisers cannot influence chatbot responses and that it will not share user conversations with advertisers. Ads will not appear next to content about sensitive or regulated topics, and users under 18 will not see ads.

CEO Sam Altman said the ad rollout aims to help OpenAI keep ChatGPT affordable. He acknowledged past criticism of ads but said this approach is necessary to support a growing number of users who want frequent access to AI without paying high fees. Altman cited Instagram’s ad experience as an inspiration for the format OpenAI will use.

ChatGPT Plus and Pro users will continue to see an ad-free experience. OpenAI said it will monitor feedback and make adjustments as the ad trial progresses. The company reported strong adoption of ChatGPT Go in earlier markets and expects the global launch to accelerate usage further.

In related news, OpenAI is reportedly planning AI-powered SweetPea earbuds, positioning them as a potential alternative to Apple’s AirPods.

For more daily updates, please visit our News Section.

Stay ahead in tech! Join our Telegram community and sign up for our daily newsletter of top stories!

(Source)

The post ChatGPT Go goes global: OpenAI bets on ads to make AI affordable appeared first on Gizmochina.

BOYA Notra is an AI-powered portable note taker built to capture, transcribe, and organize real-world conversations with ease. Developed by BOYA, it targets professionals, students, journalists, and content creators who rely on accurate records from meetings, lectures, interviews, and calls. Its core focus is simplicity, instant recording combined with intelligent processing in a pocket-sized form.

Price, Discount & Availability

BOYA Notra is launching with a 26% Super Early Bird discount, bringing the price down to $111 USD for a limited period from January 16 to January 22, 2026. Users who subscribe early can also receive an additional $10 discount code and 45 days of complimentary Premium Plan access. The device is available exclusively through BOYA’s official online store. At launch, AI transcription supports English and Japanese, with more languages and the ASK AI smart search feature rolling out via app updates in early February 2026.

All-in-One Recording Capability

BOYA Notra is the first pocket-sized recorder to fully support three daily audio scenarios: ambient room recording, handset phone calls, and Bluetooth headset calls. This eliminates the need for external microphones or multiple recording apps. A simple slide-switch operation enables instant recording, making it reliable for fast-paced environments like interviews and briefings.

Professional-Grade Audio Capture

The device features dual premium MEMS microphones with 360° omnidirectional pickup. AI noise reduction suppresses background noise by up to 30 dB, ensuring clear voice capture across rooms, phone calls, and Bluetooth headsets. The dual-microphone array helps maintain voice accuracy, improving transcription results without manual adjustments.

AI Transcription and Smart Search

BOYA Notra converts recordings into structured summaries, actionable insights, and organized notes. It supports professional templates, mind maps, and to-do lists. Powered by leading AI models, the ASK AI feature allows users to search recordings by voice or text, helping locate key moments instantly.

Storage, Battery, and Portability

With 64 GB of built-in storage, BOYA Notra can store up to 8,000 hours of audio and includes free cloud sync. It offers up to 24 hours of continuous recording and up to 365 days of standby time. The compact magnetic design fits easily into a jeans coin pocket, making it ideal for everyday carry.

BOYA Notra’s headline feature isn’t hype—it’s the first pocket recorder that genuinely covers all three daily audio scenarios: room ambience, handset calls and Bluetooth-headset calls. One thumb-sized device replaces the usual “phone + app + mic” juggle, so you’re always one slide-switch away from evidence-grade audio.

Ambient mode is the one I leave on during briefings and lectures. Drop it on the table, no gain-staging, no missing the speaker at the far end of the room. The dual-mic array keeps voices in phase so the transcript engine doesn’t hallucinate half-sentences. For reporters or students who’ve lost a quote to a muffled phone memo, that “it-just-works” safety net is worth the price of admission alone.

BOYA Notra is driven by an advanced AI framework that integrates leading large language models, including ChatGPT, Gemini, Claude, and Grok. Its ASK AI feature allows users to search their personal recording library by voice or text, instantly surfacing key moments without the need to re-listen. All recordings and transcripts can be reviewed, edited, translated, and exported in a wide range of formats such as MP3, WAV, DOCX, PDF, TXT, Markdown, and JPEG.

Designed for Security and Simplicity

Built for effortless portability, Notra features a sleek, ultra-portable magnetic design. With 64 GB of built-in storage (capable of holding up to 8,000 hours of audio), plus free cloud sync. Its 24-hours continuous recording, and a 365-day standby batter keeps the information seamlessly accessible across devices and ready to capture any moment.

From January 16-22, 2026, customers get 26% off of Boya Notra at a Super Early Bird price of $111 USD, and receive 45 days of complimentary Premium Plan access. Additionally, users who subscribe from now until January 16 will receive an extra $10 discount code.

For more information, please visit the official website here.

The post One AI Pocket Recorder for Calls, Meetings, and Notes: BOYA Notra Launches at Discount appeared first on Gizmochina.

OpenAI’s long-rumored move into consumer hardware is starting to feel very real. According to new reports from China, the company is developing an AI-powered earphone project codenamed “Sweet Pea,” which could eventually challenge Apple’s AirPods.

The hardware is reportedly being designed by Jony Ive, Apple’s former chief design officer and the man behind some of the most iconic products of the iPhone era. His involvement alone gives the project a level of credibility that most AI gadgets lack.

OpenAI acquired Ive’s design firm, io hardware, in May 2024 for $6.5 billion, and Sweet Pea could become the first major outcome of that acquisition.

A new form factor for earbuds

Unlike traditional true wireless earbuds that sit inside the ear canal, Sweet Pea reportedly features a pebble-shaped metal main unit paired with two capsule-like components that rest behind the ear. This unconventional design allows more space for high-performance chips and onboard AI computing.

At the heart of the device is a custom 2nm processor capable of handling most AI tasks locally. Instead of sending every request to the cloud, Sweet Pea would process voice commands, contextual understanding, and user interactions directly on the device.

The report also suggests that Sweet Pea could replace many basic phone functions, allowing users to control apps and services entirely through voice. To enable this, the device would include microphones and cameras that continuously perceive the surrounding environment.

Because of this constant awareness, users wouldn’t need to say a wake phrase like “Hey.” Instead, the AI would proactively offer suggestions and assistance based on what it sees and hears.

OpenAI is reportedly targeting a September 2028 release for Sweet Pea, with a first-year shipment goal of 40 to 50 million units. That’s an ambitious number for a brand-new product category, but it highlights the company’s confidence in the device.

The only problem is that 2028 is still a long way off—and a lot can change before then.

(Via 1, 2)

The post OpenAI plans AI-powered “SweetPea” earbuds as an AirPods alternative appeared first on Gizmochina.

AI has become one of the biggest selling points of Galaxy smartphones over the past few years. In fact, all the launch events for the last two generations of Galaxy S and Galaxy Z series flagships focused heavily on how these devices aim to make life easier through AI-driven features. On the other hand, the list of hardware improvements has been relatively short in those releases. This has also raised concerns about Samsung potentially charging for the very features that were highlighted as the key selling point of these devices.

The brand initially promised to keep Galaxy AI features “complimentary through 2025,” a statement that didn’t clearly explain what would happen beyond that period. It made headlines and, quite reasonably, sparked concerns among Galaxy fans who were convinced by, and grew to rely on, these features.

Now, Samsung has finally clarified its future plans regarding the key selling point of its phones. The Galaxy AI landing page has been updated with a footnote stating: “Galaxy AI basic features provided by Samsung are free. Future releases may include enhanced features or new services that are offered on a paid basis. Different terms may apply for AI features provided by third parties.”

Samsung hints at plans to charge for some advanced AI features

Samsung is likely to push more advanced — and possibly more convenient — AI features going forward that make Galaxy phones smarter. However, the statement also hints that the brand plans to eventually charge for some advanced and potentially niche AI features. The upcoming Galaxy S26 series, for instance, is expected to focus heavily on proactive, on-device AI.

“Galaxy AI basic features are those services listed under ‘Advanced Intelligence’ in the current Samsung Services Terms and Conditions,” the company further explained in the footnote.

Under “Advanced Intelligence” in Samsung’s Terms, the basic, free Galaxy AI features include communication tools such as Live Translate, Interpreter, Chat Assist, and Transcript Assist, along with productivity-focused features like Note Assist, Writing Assist, Browsing Assist, Now Brief, and Photo Assist.

While the “basic” Galaxy AI features are now confirmed to remain free indefinitely, Samsung’s decision to potentially charge for future AI capabilities could signal a broader industry shift. After all, it’s common for smartphone brands to take notes from each other’s playbooks. How the “AI features” category evolves by the end of the year will be worth watching.

It’s also worth noting that some features powered by Google Gemini, such as Gemini Live, fall under the “AI features provided by third parties” category, meaning different terms could apply to them.

Don’t miss a thing! Join our Telegram community for instant updates and grab our free daily newsletter for the best tech stories!

For more daily updates, please visit our News Section.

(Source | Via)

The post Samsung confirms basic Galaxy AI features will remain free forever appeared first on Gizmochina.

Buying a robot can cost as much as a luxury car. Chinese Robot rental firm Qingtian Rent wants to change that. Launched in Shanghai in late December 2025, the platform lets businesses rent robots by the day, just like booking a car or a food delivery rider. The idea is simple: turn robots from expensive machines into easy-to-use services.

Backed by Big Investors

Investors clearly like the idea. Qingtian Rent has raised tens of millions of dollars in seed funding, led by Hillhouse Capital, with Fosun and other major funds joining in. The money will be used for nationwide expansion, better platform tech, and a standardized service system that includes insurance and trained operators.

Explosive Early Demand

The response has been fast and loud. In just three weeks, more than 200,000 users signed up. Daily rental orders already exceed 200 and are still climbing. Brands like Haidilao, Mixue Ice Cream, and Meiyijia are using robots to attract customers, perform dances, write calligraphy, and create viral moments in stores.

Why Now? Timing Is Everything

Robot rentals are booming in China, especially ahead of the Spring Festival. After humanoid robots went viral at concerts and on TV, demand surged, and supply quickly tightened. Rental prices have also dropped sharply, making robots accessible to smaller businesses for the first time.

More Than Entertainment

While robots are popular for shows and promotions today, Qingtian Rent is betting on a bigger future. As costs fall, robots are expected to move into retail, tourism, factories, and warehouses. The company’s long-term vision is bold: to become the “Didi of robot leasing” and build a nationwide robot service network.

Robot Rentals Outside China

Robot rentals are available in other countries, but mainly in limited and specialized forms. In Japan, the US, and parts of Europe, robots are typically rented for events, exhibitions, security roles, or factory automation. These services are often run by small providers or offered as industrial leasing programs, making them business-focused and far from a mass-market, on-demand experience.

Why China Is Moving Faster?

China’s advantage lies in scale and platform thinking. Strong investor and government support, a huge domestic market, and rapidly falling robot costs have made nationwide Robot-as-a-Service platforms possible. In many other countries, robots are still viewed as long-term industrial investments rather than short-term services, slowing experimentation and delaying widespread robot rental adoption.

Read More:

• New 7-Seat Tesla 2026 Model Y Launches in US, but Families May Still Look Elsewhere
• Huawei Returns as China’s No. 1 Smartphone Brand After Five Years, Overtaking Apple
• Honda Changes Its Iconic Logo After Decades: New ‘H’ Debuts for EV Era

(via)

The post Rent a Robot for Just $25 (₹2000) a Day: World’s First Robot Rental Platform, Raises Millions appeared first on Gizmochina.

Humanoid robots crossed a commercial threshold in 2025, and 2026 will be the year the industry begins to scale, according to Counterpoint Research. The firm estimates that around 16,000 humanoid robots were installed globally in 2025, with China accounting for more than 80% of deployments, underscoring the country’s early dominance.

The market is already consolidating fast. Counterpoint notes that the top five manufacturers control 73% of the global share, four of them Chinese. AGIBOT leads with 31%, supported by over 5,000 units produced and active deployments in hotels, entertainment, and industrial logistics. Unitree Robotics follows with 27%, leveraging cost-efficient designs and rapid go-to-market execution. UBTECH, Leju Robotics, and Tesla each hover around the 5% mark, with Tesla expected to gain momentum as Optimus production ramps.

Looking to 2026, Counterpoint expects a shift from early commercialization to capacity expansion and cost compression. Larger factories, falling component costs, and growing adoption of Robots-as-a-Service (RaaS) models are set to lower barriers for enterprise customers.

By 2027, Counterpoint forecasts the global installed base will exceed 100,000 units, with logistics, manufacturing, and automotive sectors absorbing 72% of new annual capacity, positioning humanoid robots as a meaningful force in global industrial productivity.

Humanoid Robot Market Snapshot (2025)

Read More:

• Honda Changes Its Iconic Logo After Decades: New ‘H’ Debuts for EV Era
• Xiaomi Mi 17 Ultra Update Brings New Zoom Ring Leica Camera Features and Charging Fixes
• New 7-Seat Tesla 2026 Model Y Launches in US, but Families May Still Look Elsewhere

(via)

The post Why 2026 Will Be the Year of Humanoid Robots, Counterpoint Report appeared first on Gizmochina.

Following a global backlash over Grok-generated images, particularly those involving non-consensual, sexually explicit, or misleading content, X swiftly restricted its image generation features to paid, verified users. The company positioned the change as a safety measure, arguing that paid access would reduce misuse, improve accountability, and facilitate the identification of bad actors.

While the intent appeared corrective, the decision leaned on an assumption X itself has struggled with in the past: that charging users creates better behavior. History on the platform suggests otherwise.

Old Twitter verification misuse shows the flaw

Even before AI tools were part of the platform, Twitter’s verification system had credibility issues. The blue tick was originally meant to confirm identity, but over time it became more of a status symbol than a trust indicator. Verified accounts were frequently involved in misinformation, harassment, and impersonation, while enforcement remained inconsistent.

When verification later became a paid feature, the problem became more visible. Fake brand accounts, misleading profiles, and coordinated impersonation campaigns surfaced rapidly, all carrying the same blue tick meant to reassure users. The lesson from this period is clear. Verification alone does not regulate behaviour, especially when deployed at scale. Applying the same logic to AI image generation repeats a familiar mistake.

Scammers already pay, cost is not a deterrent

A common assumption behind paywalled AI tools is that bad actors will hesitate to spend money. In practice, scammers have long treated platform fees as a routine business expense. On X, spam networks, crypto fraudsters, and impersonation accounts already pay for verified access because the reach and perceived legitimacy make it worthwhile.

For those misusing AI image tools, the economics are similar. A subscription fee is insignificant compared to the potential visibility or impact of abusive content. While paid access may reduce casual misuse, organized abuse is rarely price-sensitive.

Why this approach scales badly for X

X operates at a global scale where AI-generated images can spread within minutes. Effectively addressing misuse at this level requires strong detection systems, consistent moderation, and clear enforcement policies, not just account verification.

As AI tools become more powerful and easier to exploit, relying on payment as a safeguard becomes increasingly fragile. More importantly, it risks normalising the idea that safety is optional or premium. History shows that without deeper structural controls, paywalls may slow abuse, but they do not stop it.

For more daily updates, please visit our News Section.

Stay ahead in tech! Join our Telegram community and sign up for our daily newsletter of top stories.

The post X’s paid verification won’t stop AI abuse — history proves it appeared first on Gizmochina.

The global humanoid robot market moved into a clear expansion phase in 2025, according to a new industry assessment by Omdia. Worldwide shipments reached around 13,000 units during the year, reflecting growing commercial demand beyond research labs and pilot projects. Omdia expects this momentum to accelerate sharply over the next decade, with annual shipments projected to climb to nearly 2.6 million units by 2035 as robots become more capable and cost-effective.

Among all manufacturers, AGIBOT emerged as the market leader in 2025. The company shipped more than 5,100 humanoid robots during the year, securing roughly 39% of the global market. This placed AGIBOT at the top in both shipment volume and overall market share, ahead of competitors such as Unitree, UBTech, and several emerging robotics firms.

Market share estimates show Unitree holding the second position with around 32%, while the remaining share is spread across companies, including UBTech, Leju Robotics, EngineAI, and others. Smaller contributions also came from well-known names such as Tesla and Agility Robotics, highlighting a crowded but rapidly evolving competitive landscape.

AGIBOT’s product lineup spans full-size humanoid robots, smaller half-size models, and wheeled embodied intelligent robots. These systems are already being used in practical settings such as hospitality, industrial manufacturing, logistics operations, security patrols, and data collection for AI training. Education and scientific research are also emerging as important use cases.

In Omdia’s technical evaluation of general-purpose embodied robots, AGIBOT received advanced ratings across most key categories, including mobility, manipulation, AI learning, and scalability. Based on these results, Omdia places AGIBOT in the global first tier of humanoid robot developers, alongside companies like Unitree and Tesla, all of which are pushing toward more adaptable, general-purpose robotic intelligence.

Read More:

• World’s First Humanoid Robot Cinema Worker Goes Live in China
• AI Pets Are No Longer Sci-Fi: Cocomo and INU Robot Pets Arrive at CES 2026
• AGIBOT Hits 5000-Unit Milestone, Fueling China’s Humanoid Robotics

The post Chinese Robot Company AGIBOT Took 39% of the Global Humanoid Market in 2025 appeared first on Gizmochina.

The world’s first fully autonomous humanoid cinema worker has started working at the K11 ART HOUSE cinema in Shenzhen, China. The robot, named Atom, is developed by Yuejiang Robotics and is designed to work in real public spaces, not just inside research labs.

The company says Atom can work for up to 14 hours a day, sell more than 1,000 cups of popcorn, help customers, and even handle small emergencies on its own. It is meant for busy service environments where consistency and patience matter.

Atom is built to move and use its hands in a human-like way. It runs on a special neural-driven AI “brain”, which helps it understand situations, plan actions, and make decisions without human control. The robot uses two camera-based ‘eyes’ to judge depth and distance, similar to how humans see. Its five-finger hands are controlled by a real-time feedback system, which allows the fingers to adjust grip strength smoothly, helping it hold popcorn cups, trays, and packets without slipping or crushing them.

Around the world, other well-known humanoid robots are also working or being tested, such as Pepper in customer-service roles, Sophia for social interaction, and research robots like Atlas and NASA’s Valkyrie, which focus on balance and movement.

Atom’s deployment also raises an important question: will humanoid robots replace human workers? Experts say that, for now, robots are more likely to handle repetitive, tiring, or risky tasks, while humans continue to lead in creative, emotional, and decision-based work, making the future more about human-robot teamwork than replacement.

Read More:

• AI Pets Are No Longer Sci-Fi: Cocomo and INU Robot Pets Arrive at CES 2026
• World’s Thinnest OLED TV Announced by LG: It Is Thinner Than a Pencil, And Wireless
• China is the World’s Robot Superpower: Installs More Robots Than the Rest of the World Combined

(via)

The post World’s First Humanoid Robot Cinema Worker Goes Live in China appeared first on Gizmochina.

As artificial intelligence continues to weave itself into everyday life, a new category of technology is quietly reshaping how humans think about companionship: AI pets. For many, the idea evokes memories of iconic fictional companions like BB-8 from Star Wars, a droid that felt less like a machine and more like a loyal friend.

BB-8’s personality was intentionally designed to mirror the charm of a curious puppy, reinforcing how deeply emotional bonds with robots can resonate with people. Today, real-world companion robots are beginning to chase that same sense of warmth, presence, and connection.

At CES 2026, this shift was on full display as Japanese startup Ludens AI introduced two deeply expressive companion robots: Cocomo and Inu, signaling a world where digital companionship may soon feel as natural as living with a pet.

Ludens AI’s philosophy is refreshingly different from the growing trend of “smart helpers.” Instead of building robots that clean or organize, the company focuses on machines that listen, react, and bond emotionally over time. Their message is simple: AI companions don’t need jobs, they just need to exist alongside people.

Cocomo: A Warm, Mobile Robot Pet Designed for Emotional Bonding

Cocomo, the star of Ludens AI’s showcase, is positioned as an autonomous robot pet that follows its owner around the home. With a wheeled chassis, ear-like decorations, and an orange, teddy-bear-like exterior, Cocomo feels intentionally playful and approachable. Its body temperature is carefully maintained near human warmth, around 37°C in normal use and up to 39°C during frequent touch, eliminating the cold, mechanical feel often associated with robots.

Instead of speaking, Cocomo communicates through humming and soft sounds, reinforcing its pet-like character. Over time, it learns what makes its owner happy, what brings comfort, and when to offer spontaneous gestures or gentle surprises. Through voice, touch, and movement-based interactions, Cocomo is designed to create a slow-growing emotional bond rather than instant novelty. Ludens AI describes it as a companion meant to evolve alongside everyday life, a robot that remembers, imitates, and adapts.

Inu: A Minimalist “Desktop Alien Puppy” for Workspaces

In contrast, Inu is a smaller, stationary desktop companion created for work environments. Marketed as a “desktop alien puppy,” it reacts to voice, touch, and nearby sounds with playful tail wags and expressive blinking eye animations. Inu doesn’t move around rooms; instead, it offers light emotional presence while users work, serving as a gentle, comforting desk companion rather than a task-oriented assistant.

Both Cocomo and Inu highlight Ludens AI’s belief that robots don’t need to be useful in a traditional sense; they simply need to be present, warm, and emotionally engaging. With crowdfunding campaigns planned later this year, these companions may soon enter real homes and offices.

A New Kind of Emotional Technology

The debut of Ludens AI’s robots signals a broader cultural shift, from machines as tools to machines as companions. As emotional AI continues to develop, Cocomo and Inu suggest a future where companionship may extend beyond pets and people, into a new world of soft-spoken, expressive, and deeply personal robotic presence.

Read More:

• China Sells Humanoid Robot for the Price of an iPhone
• Humanoid robot’s hilarious mishap goes viral: kicks trainer and mimics the pain
• AI Learns to Pilot a Space Robot : Navigates the International Space Station Faster

(via)

The post AI Pets Are No Longer Sci-Fi: Cocomo and INU Robot Pets Arrive at CES 2026 appeared first on Gizmochina.

Training large AI models has become one of the biggest challenges in modern computing—not just because of complexity, but because of cost, power use, and wasted resources. A new research paper from DeepSeek proposes an approach that could help ease some of those pressures.

The method, called manifold-constrained hyperconnection (mHC), focuses on making large AI models easier and more reliable to train. Instead of chasing raw performance gains, the idea is to reduce instability during training—a common issue that forces companies to restart expensive training runs from scratch.

In simple terms, many advanced AI models fail mid-training. When that happens, weeks of work, massive amounts of electricity, and thousands of GPU hours are lost. DeepSeek’s approach aims to prevent those failures by keeping model behavior more predictable, even as models grow larger.

This matters because AI training today consumes enormous amounts of power. While mHC doesn’t make GPUs themselves use less power, it can reduce wasted power by helping models finish training without crashing or needing repeated restarts.

Another benefit is efficiency at scale. When training is more stable, companies don’t need to rely as heavily on “brute force” methods—such as throwing more GPUs, more memory, or longer training schedules at a problem just to make it work. That can lower the total energy used over the full training process.

DeepSeek’s research doesn’t claim to solve hardware shortages or energy challenges overnight. Instead, it represents a quieter but important improvement: making better use of the resources already available. Over time, techniques like this could help AI developers train powerful models with fewer wasted compute hours and lower overall energy consumption.

As language models continue to grow, reducing inefficiency may become just as important as chasing higher performance—and that’s where DeepSeek’s new AI architecture could make a real difference.

Don’t miss a thing! Join our Telegram community for instant updates and grab our free daily newsletter for the best tech stories!

For more daily updates, please visit our News Section.

(Sources: 1, 2 | Image)

The post DeepSeek kicks off 2026 with new AI architecture aimed at more efficient model training appeared first on Gizmochina.

A training session involving Unitree’s G1 humanoid robot has gone viral for all the wrong—and funny—reasons. The clip, first uploaded to China’s Bilibili on December 25, shows a human trainer controlling the G1 using a motion-capture suit. The setup allows the robot to mirror the trainer’s movements in real time, a common method used to teach humanoid robots complex actions.

Things are going smoothly at first. The trainer throws a few punches and kicks, and the G1 follows along without issue. But during one sequence, the robot appears to misjudge a rotation and delivers a kick straight into the trainer’s groin.

The reaction is immediate. The trainer doubles over, clutching himself before dropping to his knees. Seconds later, the robot copies the same posture, bending forward as if reacting to the pain itself—a moment that has turned the clip into a meme across social platforms.

The video quickly spread beyond Bilibili, appearing on X (formerly Twitter) the next day. It was shared widely, including by Tesla Cybertruck chief engineer Wes Morrill, and has since picked up traction on Reddit and other forums.

The humanoid robot has previously been shown playing table tennis, shooting basketballs, and performing martial arts routines. This incident, however, highlights one of the risks of close-range teleoperation, where small timing errors or misread movements can lead to unexpected results.

Despite the mishap, teleoperation remains an important tool for training humanoid robots, especially for tasks that are too complex to pre-program. Still, the clip serves as a reminder that even advanced robots are far from flawless.

For now, the G1’s accidental kick—and its perfectly timed imitation of human suffering—has given the internet a good laugh.

Don’t miss a thing! Join our Telegram community for instant updates and grab our free daily newsletter for the best tech stories!

For more daily updates, please visit our News Section.

(Source)

The post Humanoid robot’s hilarious mishap goes viral: kicks trainer and mimics the pain appeared first on Gizmochina.