Artificial Intelligence news in 1-minute reads — fast and easy to keep up with
Last updated: 21/08/2026, 16:00
Sanjay Mehrotra, CEO da Micron, afirmou que a inteligência artificial (AI) transformou profundamente o ciclo de demanda no setor de memória. Segundo Mehrotra, a dependência de memória por parte dos sistemas de IA criou uma demanda mais duradoura, mudando a dinâmica tradicional do setor, que antes seguiu um padrão de 'aumento da demanda → aumento da produção → excesso de oferta → queda de preços'. A declaração foi feita enquanto a Micron constrói uma nova unidade de fabricação de semicondutores em Boise, no Idaho, que incluirá dois fábricas, cada uma com área equivalente a 10 campos de futebol americano. A unidade prevê-se iniciar as operações em 2027. Mehrotra destacou que a demanda por memória não se limita aos centros de dados, mas também inclui setores como automação, robótica e dispositivos de consumo inteligentes. No entanto, ele admitiu que a produção atual da empresa não é suficiente para atender à demanda, já que os clientes finais estão comprando todos os produtos disponíveis e os clientes de centros de dados desejam uma quantidade duas vezes maior do que a que a Micron pode oferecer.
Bloomberg reports that NVIDIA has initiated preliminary discussions with the South Korean AI chipmaker Rebellions on a potential collaboration. The talks took place during a meeting between the leaders of both companies at NVIDIA’s headquarters in Santa Clara, California. According to sources, the discussions are in an early stage, with possibilities including technological cooperation, investment, or even an acquisition, though the deal could still fall through. Rebellions specializes in NPUs for AI inference acceleration, with its latest product, the Rebel100, offering 2 PFLOPS of FP8 computing power, 523MB of SRAM cache, and 144GB of HBM3E memory.
Kakao, a major South Korean IT company, has approved a restructuring plan during a board meeting, which will split the company into two new entities: Kakao AI and Kakao X. The split, based on net asset value, will allocate 36% to Kakao AI and 64% to Kakao X. The move is intended to adapt to the AI era by reorganizing capital allocation. Kakao AI will focus on developing AI agents, with KakaoTalk and Kanana as its core platforms. The new entity aims to integrate Kakao’s extensive services to create unique user experiences and business models. Users will be able to issue commands via KakaoTalk, enabling AI agents to perform tasks such as searching, purchasing, and paying. The company targets a 50% increase in user engagement and a 20% annual revenue growth by 2030, aiming for a revenue of approximately 6 trillion won (around 291.06 billion RMB). Kakao X will manage existing subsidiaries, including Kakao Bank, Kakao Pay, and SM Entertainment, while exploring new areas such as virtual assets and investing in innovative startups. The company plans to hold a special shareholders’ meeting on December 17 and complete the split by January 1, 2027. Kakao AI is scheduled to re-list on the stock market on January 27, 2027, alongside a name change and potential listing for Kakao X.
A OpenAI anunciou que o modelo GPT-Image-2 agora suporta a geração de imagens com fundo transparente. Os utilizadores podem criar imagens no formato PNG com um canal alfa, permitindo uma melhor apresentação em contextos como design de páginas de produtos em e-commerce, ilustrações para apresentações, elementos de adesivo e imagens de produtos. A funcionalidade está disponível apenas através de API, e os utilizadores devem ativar-a ao incluir o parâmetro background=transparent. A OpenAI aconselha os utilizadores a não descreverem o fundo nos prompts e a verificar cuidadosamente as imagens geradas para garantir a precisão dos números presentes nelas.
Qwen has introduced new capabilities within its work task mode, including skills, connectors, and work partners. These features are now available in the desktop version of Qwen, accessible through the sidebar under the 'Skills · Connectors · Partners' section. Over 200 skills and connectors are already available, with the option for users to create custom skills by organizing common workflows, standards, and templates into reusable processes. Connectors allow direct information retrieval and task processing within commonly used tools, while work partners enable users to select specialized agents based on their needs, forming collaborative teams to tackle complex tasks. The new features can be used independently or combined as needed, allowing users to standardize their workflows and enhance productivity.
An AI training startup, Micro1, has submitted a $12.5 million bid for Spirit Airlines' data, aiming to challenge Google's winning $10 million offer. The bid, made on Wednesday, comes after Google successfully secured the data in a recent auction. Micro1's CEO, Ali Ansari, stated that the company's offer is significantly higher, reflecting the value of real-world data for training AI models. Spirit Airlines is currently in bankruptcy court, selling off its assets, including vast amounts of corporate data, following its collapse last year. Bankruptcy experts noted that while it is unusual for a late bid to be considered, it is not impossible. Ansari emphasized that realistic data is crucial for AI models to learn how to operate in real-world environments, calling Google's bid 'actually quite low' given Spirit Airlines' long history. Micro1, based in San Francisco, also pays gig workers to enhance AI models for its clients and runs digital environments where AI agents can practice tasks like booking train tickets. The company, founded in 2022, has around 150 employees and has received offers at a $2.5 billion valuation, according to Forbes. Another AI training startup, Mercor, also bid for the data, initially offering $5 million, which was later increased to $7.5 million, still falling short of Google's final bid. Google, Spirit Airlines, and Mercor have all declined to comment on the matter.
DeepReinforce has announced the release of the Ornith-1.5 series of open-source AI models, emphasizing the integration of a 'self-improvement loop' during training. This mechanism allows the models to autonomously enhance their performance by continuously generating more challenging tasks during the learning process. The largest model in the series, Ornith-1.5-397B, is based on a mixture-of-experts (MoE) architecture with 397 billion parameters, and its performance surpasses that of Claude Opus 4.8 in certain tests. Another variant, Ornith-1.5-35B-A3B, also uses a MoE architecture with 350 billion parameters, outperforming dense models such as Muse-Glimmer-30B and Gemma-4-31B. The Ornith-1.5-9B model, a dense architecture with 9 billion parameters, demonstrates superior performance in most tests compared to the parameter-heavy Gemma-4-31B. Additionally, a quantized version named Ornith-1.5-9B-Mobile is available for mobile devices, including Android phones and the upcoming iPhone 17.
Jiajia Vision, a physics AI company founded in 2023, has unveiled its general-purpose bipedal humanoid robot, Maker L01. The robot features 31 joint motors and is capable of performing actions such as running, jumping, and flipping, with a maximum movement speed of 4 meters per second. It was demonstrated in a table tennis scenario. Weighing 42 kilograms, the robot is equipped with a 518Wh hot-swappable battery and is designed to be fully open-source, supporting both research and competition development out of the box. Jiajia Vision claims to be the first company in China focused on 'world models,' a core AI framework used to simulate environmental dynamics and predict future states. These models are expected to play a key role in embodied intelligence and autonomous driving. The company recently completed a billion-yuan A1 round of financing in November 2025, with investment from a well-known industry player and Huakong Fund. According to Qiushiwei, the 'well-known industry player' is Huawei's Habor investment arm. Jiajia Vision's product lineup includes the world model platform GigaWorld, the embodied foundation model GigaBrain, and general-purpose embodied hardware such as Maker. The company's focus on world models positions it at the forefront of physical AI development.
On August 20, during the 2026 World Robot Conference (WRC 2026), the Beijing Humanoid Robot Innovation Center officially launched the lightweight humanoid robot Tian Gong Omni. The robot integrates hardware and edge intelligence, marking the first globally available Humanoid Native open platform that combines perception, mobility, and full-body control. The Tian Gong Omni stands at 1.35 meters tall and weighs 39 kilograms, with the capability to handle moderate loads. It is designed for use in narrow spaces, emergency rescue, and home services. The platform also opens its secondary development system, allowing developers to access joints, sensors, systems, and control APIs, thereby reducing development costs and accelerating innovation. CEO Xiong Youjun of the Beijing Humanoid Robot Innovation Center stated that the development of embodied intelligence is still in its early stages, and entering the market is not too late. He emphasized that Tian Gong Omni is primarily targeted at industrial applications and will not directly compete with companies such as Songyan Dynamics and Jiasu Jihua. Instead, the focus is on providing a standard platform for industry-wide secondary development, promoting common and core technologies. Xiong also highlighted the importance of creating a cost-effective, scalable platform for mass production. He mentioned that future efforts will focus on the 'big brain' and 'small brain' platform, aiming to offer a flat ecosystem for the industry. Regarding the return on investment (ROI) for humanoid robots in factory settings, Xiong acknowledged that the current ROI is lower than that of industrial robots or human workers, but the gap is rapidly narrowing.
A AI chip startup, Fractile, has reached a preliminary agreement with Anthropic for a supply contract valued at approximately $250 million, according to a report by Bloomberg. The deal, which is expected to begin deliveries in 2027, involves Fractile’s AI ASIC inference accelerators. The company’s chips utilize an innovative 'analog memory computing' architecture, enabling 25 times faster performance and 10 times lower costs when running leading models. Fractile previously raised $220 million in May 2026, valuing the company at around $1 billion. The firm aims to secure additional funding at a valuation of $6.5 billion, with some of the new investment based on potential valuations below this target. The company’s technology is positioned as a competitive alternative to existing AI hardware solutions, offering significant efficiency gains over traditional approaches. The supply agreement with Anthropic highlights Fractile’s growing influence in the AI chip market, as it seeks to scale its operations and meet increasing demand from major AI firms. The deal underscores the company’s strategic focus on delivering high-performance, cost-effective solutions for large-scale AI deployment.
Yu Shi Technology, the first publicly traded company in China's humanoid robot sector, saw its stock price drop by 18.7% on the second day of its listing on the Sci-Tech Innovation Board. The company's shares opened at 1,100 yuan, a 629.44% increase from the issuance price of 150.80 yuan, pushing the total market capitalization to as high as 444.9 billion yuan. However, by the end of the trading session on August 19, the stock closed at 845 yuan, with a total market value of over 341.8 billion yuan. The following day, the stock fell sharply, ending at 277.9 billion yuan in market value, a decline of 1,670 billion yuan from its peak. During the 2026 World Robot Conference, Yu Shi Technology's founder and chairman, Wang Xingxing, addressed the limited adoption of their products in industrial and domestic settings, citing inefficiencies and limitations in current capabilities. He stated that while the robots can perform certain tasks, their efficiency compared to humans remains low. Wang also predicted that humanoid robots could enter the mass market within a decade, provided manufacturers overcome current practical challenges. He emphasized that the technology would need to handle 80% of tasks through voice commands in unfamiliar environments to drive market growth, a goal he estimates could be achieved in two to three decades.
Alibaba has officially released Qwen-UI-Agent, a GUI intelligent agent foundation model designed to operate effectively in real-world environments, including mobile devices, computers, web browsers, and DeepSearch. The model demonstrates strong performance across multiple benchmarks, surpassing leading models in several categories. On mobile tasks, Qwen-UI-Agent achieved 82.1% on MobileWorld, outperforming GPT-5.6 Sol, Claude Opus 4.8, Seed 2.1 Pro by 12.0, 14.6, and 8.9 percentage points respectively. It also scored 92.2% on MobileWorld-Real, surpassing Gemini 3.1 Pro, Claude Opus 4.8, GPT-5.6 Sol, and Seed 2.1 Pro. On Android Daily, it reached 97.5%, nearly perfect. On desktop tasks, it scored 79.5% on OSWorld-Verified, outperforming GPT-5.5, Gemini 3.1 Pro, and Seed 2.1 Pro. It also achieved 40.0% on OSWorld-v2 Partial, with a 58% reduction in execution steps compared to the baseline. In web and DeepSearch tasks, it scored 73.6% on WebArena, the highest among all compared models, and 75.0% on BrowseComp-ZH. In GUI grounding tasks, it achieved 81.5% on ScreenSpot-Pro, setting new state-of-the-art records on four other benchmarks. The model also excels in general and agentic capabilities, outperforming both training base models and GUI-specific models. It operates in a real-world environment with over 100 real phones and 150 apps, using its own MobileWorld-Real benchmark with 400+ tasks and 100+ apps. The model is designed to handle complex, long-tail tasks in real-world scenarios. It integrates safety checks throughout task execution, refusing to perform any interface actions for illegal or high-risk requests and pausing for user confirmation in sensitive operations such as payments, data deletion, and privacy authorization. Additionally, it supports command-line operations and batched actions, with nearly half of the actions on desktop tasks being batched, significantly improving efficiency. The model can also perform online reinforcement learning over more than 100 steps, with approximately 10,000 concurrent environments running simultaneously. The project is available at https://tongyi-mai.github.io/Qwen-UI-Agent and on GitHub at https://github.com/Tongyi-MAI/MAI-UI.
Galaxy Universal, a company specializing in artificial intelligence, unveiled its first bipedal robot, Galbot ET1, at the 2026 World Robot Conference. The robot, named 'Xingzai,' is equipped with the Galaxy Star Brain, allowing it to interact in real-time without relying on pre-set scripts. According to Wang He, the founder and CTO of Galaxy Universal, Xingzai is the first physically intelligent agent capable of autonomous learning. During its debut, Xingzai performed a high-level street dance, including handstands and synchronization with a human dancer. Wang highlighted that the robot's movements are supported by the world's top-tier motion control models and data infrastructure, enabling it to recognize, extract, and replicate human actions in real-time. The demonstration showcased Xingzai's ability to improvise and engage in spontaneous interactions, marking a significant advancement in humanoid robotics.
The chief executive officer of HUMAIN, Tareq Amin, has revealed details about the upcoming Horizon Ultra laptop through an unboxing video. The device is set to feature the Qualcomm Snapdragon X2 Elite processor, which will serve as its core computing engine. A HUMAIN logo is also embedded on the right Alt key of the dual-language English-Arabic keyboard located on the left side of the laptop. Amin expressed particular excitement about the operating system of the Horizon Ultra, which is designed with a focus on 'intent' and is intended to support the next generation of agent-based AI. The laptop will be showcased at the LEAP 2026 conference in Riyadh, Saudi Arabia, from August 31 to September 3. The unveiling comes as part of HUMAIN's efforts under the Saudi Arabia Public Investment Fund (PIF) to advance artificial intelligence technologies. The Horizon Ultra represents a significant step in the company's strategy to develop AI-driven hardware and software solutions.
U.S. cybersecurity agencies have issued a warning that hackers are using artificial intelligence to target water systems across the country. The Federal Bureau of Investigation, the National Security Agency, and the Cybersecurity and Infrastructure Security Agency (CISA) have reported that cybercriminals are actively infiltrating Siemens S7 programmable logic controllers, which are used to manage automated processes in energy, water, manufacturing, and agriculture. The agencies emphasized that the attacks are targeting all instances of these devices, which could lead to system outages or safety risks if compromised. Hackers are leveraging AI to generate exploit scripts and identify vulnerable controllers based on publicly available information. Older systems running outdated software or lacking adequate security measures are particularly at risk. CISA has long advised critical infrastructure operators to avoid direct internet connections for these devices, noting that rural areas are more susceptible to attacks due to the extensive coverage required for their infrastructure. A cybersecurity professional highlighted that attackers are now using AI to understand how these devices operate, raising concerns about the potential for more sophisticated attacks. Despite the weak inherent security of these devices, the threat has intensified in recent months, with suspected Iranian hackers launching multiple attacks on U.S. water and wastewater facilities. Several states, including Minnesota, Michigan, Arkansas, Georgia, and New Jersey, have reported breaches of their water infrastructure. CISA has issued this latest warning in response to a series of recent cyberattacks on critical infrastructure, noting that the frequency and intensity of such attacks have increased since Iranian hackers first targeted connected systems within the country.
Ant Group's Bai Ling has officially released the Ling-3.0-tiny and Ling-3.0-flash base models, along with their pre-training and mid-training checkpoints. The release includes six checkpoints, covering pre-training, mid-training, and WSM (Warmup-Stable and Merge) stages. These checkpoints are designed for continuous pre-training, domain-specific fine-tuning, preference optimization, reinforcement learning, distillation, and research into long-context and MoE systems. The base models are not yet aligned with instruction-following tasks and are not recommended for direct deployment as chat services or in safety-critical applications without further training and evaluation. The Ling-3.0-tiny-base model has 7.9 billion total parameters and 1.3 billion activation parameters, achieving better results than its predecessor despite a 50% reduction in total parameters. The Ling-3.0-flash-base model, with 124 billion total parameters and 5.1 billion activation parameters, offers greater capacity and sparse activation design, making it suitable for real-world applications. Both models demonstrate strong performance in coding, complex reasoning, and long-context tasks, even when compared to models with twice or thrice their parameter count. The models are available on Hugging Face and ModelScope, with links provided for each checkpoint version.
Adobe has expanded its Firefly AI creative studio services by adding three new audio capabilities, now available to all users. The update allows creators to generate music, voiceovers, and sound effects within a single workspace, eliminating the need to switch between multiple applications or browser tabs. The new features are designed to support content creation for social media, live video logs, short films, product tutorials, and podcast segments. The Generate Music tool is powered by the Firefly Music Model and can create original compositions based on the duration and mood of a video. Adobe emphasized that the generated music is safe for commercial use and comes with a general license. For Generate Speech, Adobe offers two options: the Firefly Speech Model and the ElevenLabs model. This feature converts text scripts into voiceovers, with the ability to adjust voice, speech rate, and tone. The Generate Sound Effects tool is driven by the Firefly Audio Model and can produce custom sound effects based on the actions, rhythm, and atmosphere within the content. These tools aim to streamline the creative process for users working with AI-driven audio production.
Google announced on August 20 that its open-source Gemma model family has surpassed 1 billion downloads. Over the past two years, developers have created more than 100,000 variants of the Gemma models, forming what Google refers to as the 'Gemmaverse,' an innovative ecosystem. The company highlighted various applications of the Gemma family, including space-based image analysis, digital standardization of medical reports, innovative cancer treatment options, and animal language recognition. These models are said to be enhancing the quality of human development. Google plans to launch an 'Awesome Gemma' repository on GitHub, which will serve as an official curated directory of the best community projects, fine-tuning examples, tutorials, and developer tools within the Gemmaverse.
Waymo, a self-driving taxi company under Alphabet, has revealed its custom-built 5nm ASIC chip designed specifically for real-time processing of sensor data and advanced machine learning tasks. The chip is part of a broader computing system that delivers data center-level performance while operating within the constraints of an automotive environment. The system has seen a 20-fold increase in computational power over the past eight years, with ultra-low latency software minimizing the delay between data capture and actionable output. The system is also built for resilience, featuring dual computing engines that provide redundancy and enhanced reliability in extreme conditions. The computing architecture integrates heterogeneous components, including CPUs, GPUs, accelerators, and advanced machine learning technologies. In addition to sourcing off-the-shelf components, Waymo has developed its own custom chip to further enhance system performance. The 5nm ASIC is designed to process raw data streams from LiDAR, radar, and cameras, extracting critical information and feeding it into the inference engine for sensor fusion machine learning models. The chip itself provides over 1000 TOPS of machine learning compute power, dedicated to front-end processing and machine learning tasks.
Bloomberg reported on August 20 that Anthropic plans to revise its data retention strategy, which was previously implemented to enhance cybersecurity. The company will now allow enterprise customers to retain data generated by its advanced AI models within their own cloud infrastructure rather than Anthropic’s facilities, for a period of 30 days. This updated system is expected to launch later this year. Anthropic introduced the 30-day data retention policy in June as part of its efforts to improve cybersecurity, ensuring that all data traffic from enterprise clients using its cutting-edge models is stored for potential investigation into cyberattacks. The company emphasized that retained customer data will not be used for training its own AI systems, but the policy faced criticism from some clients, raising concerns about data privacy and potential shifts to competitors offering a 'zero retention' approach. The proposed change aims to alleviate these concerns by giving customers more control over their data storage options.
