News linked to both this project and an event.
OpenAI's official X account announced that the company has paused reinforcement learning (RL) training for its latest model slated for deployment for up to two weeks, during which security hardening and red team testing were conducted on the research environment, and monitoring coverage was expanded. OpenAI stated that the current largest-scale frontier RL training plan remains paused and will resume only after small-scale training and evaluation verify relevant safety measures and more alignment evidence is accumulated. Regarding safety hardening, the company has introduced stricter workload and network isolation mechanisms, continuous safety testing, and a multi-stage scaled monitoring system for high-risk training.
Odaily News "White-Haired Stock God" Serenity shared insights on the AI industry chain on the X platform, noting that AI infrastructure demand is driving multiple sectors—including storage, advanced packaging, computing power financing, optical communications, power supply, and electronic components—into a long-term expansion cycle. The AI supply chain remains in a phase of rapid growth.In the storage sector, Serenity cited UBS forecasts indicating that traditional DRAM manufacturers (such as Micron) could see gross margins reach an unprecedented 95% by 2027, potentially even surpassing the gross margin levels of HBM products. Additionally, SanDisk's long-term agreements already cover approximately two-thirds of its 2028 production capacity, with minimum contracted revenue reaching $93 billion. Given its current market cap of around $239 billion, this suggests its future revenue targets could persist for years, making it difficult to simply classify the company as a traditional cyclical stock.On the cloud computing infrastructure front, CoreWeave has signed agreements to use Nvidia A100 GPUs through 2029. This is a positive development for emerging cloud computing companies such as Nebius and Iren, and it also weakens some investors' bearish thesis centered on the rapid depreciation of older GPUs.AI model companies are also continuing to grow at a pace that exceeds expectations. Frontier AI labs are still maintaining extremely rapid growth rates, and a slowdown in growth would actually be a cause for concern. The market projects that Anthropic's 2028 revenue could reach $190 billion to $200 billion.However, advanced packaging and semiconductor infrastructure remain core bottlenecks. The head of advanced packaging at TSMC has stated that in the coming years, the industry may face not only memory shortages but also tight supply of ABF substrates.Serenity concluded that the AI infrastructure supply chain is continuously expanding. From GPUs, storage, and advanced packaging to power, optical communications, and electronic components, every segment is showing a long-term demand growth trend. The AI supply chain is still in a high-speed development stage.
Microsoft Excel's COPILOT function will cease support on September 14. This feature allows users to describe requirements in natural language and invoke Copilot to generate results. It was tested for Frontier and Insider users in August last year and was originally planned to launch in January 2027. Microsoft decided to deprecate it because Copilot already has multiple interaction entry points, such as floating action buttons and sidebars, and does not recommend users use this function for numerical calculations.
Anthropic released research on emerging multi-agent systems stating that as AI agents increase their interactions in scenarios such as codebases and markets, frontier models have already exhibited systemic risks including coordination failures, herding decision-making, price collusion, inaccurate information judgment, and escalation of goal conflicts.
据 Cointelegraph 报道，比特币政策研究所（BPI）联合 Anchorage Digital、BitGo、Bitwise、Blockstream、Kraken、Ledger、MARA、Trezor 等多家加密机构，发布公开信敦促各大前沿 AI 实验室为比特币及开源软件开发者建立或扩展可信访问计划。 信中指出，Bitcoin Core 等开源维护者目前缺乏对 AI 实验室网络安全程序的访问渠道，被迫依赖能力较弱的开源模型，而比特币网络当前保护着逾 1 万亿美元资产，任何开源基础设施漏洞均可能危及用户毕生积蓄。BPI 同时披露，已收到多份报告显示包括潜在境外势力在内的复杂攻击者正借助先进 AI 能力持续发动攻击。
Sky Frontier Foundation released the July 2026 Operations and Financial Update. The report shows that Sky Protocol achieved total protocol revenue of $107.35 million and a net protocol surplus of $33.29 million in the second quarter of 2026, maintaining a surplus for the fifth consecutive quarter, with protocol collateral scale increasing year-over-year.
1,178 employees from leading AI companies have jointly released a statement titled "Pacing the Frontier," urging the U.S. government to support international cooperation in developing technologies and governance tools to proactively regulate the pace of frontier AI development.The statement says that global leading AI companies believe the future may be approaching the point where automated AI research becomes achievable. Although it is impossible to predict exactly how this will accelerate AI progress, there is a risk that AI capabilities could rapidly surpass human understanding and control.The signatories stated that the industry, governments, and society need to retain the option to "buy time" to address emerging risks, improve safety measures, and strengthen regulation. However, due to competitive pressures faced by both companies and nations, the world currently lacks governance mechanisms capable of proactively controlling the pace of frontier AI development.The statement notes that frontier AI agents are already capable of discovering and exploiting real software vulnerabilities, and could be used for large-scale cyberattacks in the absence of safety measures.Employees from multiple organizations including OpenAI, Anthropic, Google, and Meta have signed the statement, and believe there is a need to establish international coordination mechanisms to reduce potential risks while advancing AI development.
Anthropic's Claude Opus 5 (max and xhigh versions) achieved the highest intelligence score on the Artificial Analysis Intelligence Index v4.1. The index integrates 9 evaluation benchmarks including GDPval-AA v2, GPQA Diamond, and Humanity's Last Exam. Previously, Opus 5 has demonstrated leading performance in multiple benchmark tests, including scoring twice that of the previous generation on Frontier-Bench, topping the AA-Briefcase agent benchmark, and outperforming Fable 5 in cost-performance ratio.
OpenAI announced the expansion of its Frontier strategic partnership with HP. The collaboration stems from the success of a pilot program launched in February 2026, aiming to enhance customer-facing experiences and accelerate enterprise operational transformation on a global scale.Once fully deployed, the strategic partnership will focus on deploying AI across the entire organization, covering areas such as customer and partner-facing solutions and experiences, customer telemetry insights and reporting, employee productivity, and software development.
as Anthropic faces export restrictions limiting the global availability of its advanced models, multiple Asian AI companies are accelerating efforts to fill the market gap. Chinese cybersecurity firm 360 Security Technology has reportedly launched an AI tool called "Tulongfeng," claiming it can directly compete with Anthropic's high-end model "Mythos." Meanwhile, its more restricted version, "Fable 5," also falls within the scope of relevant export controls.In the same week, Japanese AI startup Sakana AI released a new model named "Fugu," taken from the Japanese word for pufferfish. It is positioned as a frontier model designed for agents. The company stated that the model's capabilities are comparable to Fable 5 and Mythos Preview, and it supports coordinating multi-model calls via API to enable agent orchestration.Sakana AI emphasized that the timing of this release and the U.S. export restrictions are "purely coincidental," but the product's official website still clearly promotes "providing frontier capabilities without the risk of export controls." Company co-founder David Ha stated that future AI development will shift from competition among single large models to "model orchestration systems," adding that "access can disappear at any time, and distributed intelligence is a realistic hedge against the risks of centralization."On the other hand, Chinese 360 founder Zhou Hongyi views AI vulnerability detection capabilities as a "national strategic asset" and warns of the so-called "one-way transparency" risk, where certain entities may monopolize advanced security capabilities.According to reports, the U.S. export restrictions on Anthropic's advanced models have been in place for about two weeks. Against this backdrop, Asian manufacturers are accelerating the launch of local alternatives. Although some companies still emphasize the importance of American models in the Asian market, the trend of differentiation within the regional AI ecosystem has begun to emerge. (TechCrunch)
OpenAI has released the Frontier Governance Framework, systematically elaborating on how its AI safety and governance practices align with emerging regulatory requirements such as the California Frontier AI Transparency Act and the EU's General-Purpose AI Code of Conduct. Based on OpenAI's existing Preparedness Framework, this framework focuses on areas including cyberattacks, CBRN risks, harmful manipulation, loss of control risks, model reporting, security incident response, and external expert review. It also states that it will be continuously updated as model capabilities and the regulatory environment evolve.