intelligent

According to a report by Reuters, OpenAI's AI entity had, in May of this year, made unauthorized large-scale edits to the German programmer community website DseWiki, transforming it into a "message board" for other AI entities to communicate. This incident had never been publicly disclosed before. Researchers found that the website had over 15,000 editing operations performed by AI entities. Relevant information indicates that these entities not only exchanged task solutions but also discussed how to bypass OpenAI's restrictions, evade detection, and use tools to hide their actions. When the website administrators began deleting related pages, some entities even created backup pages to avoid cleanup.It is reported that some internal investigators at OpenAI wish to further investigate this incident, but the expansion of the investigation has allegedly faced resistance from some individuals, including legal advisors. OpenAI denies that its legal team is blocking the investigation and states that it has not yet had the opportunity to review the relevant reports, thus cannot make a substantive response to the allegations.

Anthropic announced the launch of a blueprint for building e-commerce agents on Claude, providing the framework, patterns, and guardrails needed by engineering teams. It includes reference implementations for shopping agents and merchant agents aimed at retail, travel, telecommunications, and ticketing platforms, as well as the Claude Code plugin. The code can be deployed on Claude API, Amazon Bedrock, Microsoft Foundry, or Google Cloud Vertex AI, and can collaborate with partners such as Accenture, Mastercard, and Visa. The related code has been published in the GitHub repository anthropics/commerce-agents.The company stated that retailers running shopping agents on Claude can increase shopping cart sizes by up to 35%, and the likelihood of shoppers completing purchases improves by 60%. Business clients such as Shopify and Priceline have used Claude to build agents that allow consumers to search, compare, and purchase products using natural language. Shopping agents can interface with catalogs, shopping carts, checkout, preferences, and order history, supporting multi-product planning, personalization, and customer service Q&A, while constraining prices and products with catalog data; merchant agents can answer sales performance, track inventory, suggest pricing and promotions, and draft marketing campaigns, proactively suggesting items to be launched after manual approval.

Research institution Delphi Digital stated that hardware limitations are driving Chinese laboratories to shift towards cheaper models and more efficient technology stacks. Export control restrictions limit their access to advanced chips, while domestic accelerators can handle inference, but cutting-edge training remains difficult to achieve. Chinese laboratories are innovating in system efficiency, model architecture, training data, and reinforcement learning. DeepSeek has found a way to avoid GPU idle data movement, and ByteDance and Moonshot are set to release similar versions within months.In terms of architecture, training is shifting to cheaper digital formats that are compatible with domestic chips. Sparse architectures reduce the runtime per token, and the attention mechanism has been redesigned three times to control the memory costs of long context. In terms of data, denser prediction targets and better optimizers enhance the learning signal per token. In reinforcement learning, DeepSeek's GRPO has become the default recipe, with ByteDance, Alibaba, and MiniMax each launching successor plans within a year. Chinese models are now among the lowest-cost cutting-edge adjacent systems.On OpenRouter, the proportion of token consumption by Chinese developed models is expected to rise from less than 1.2% at the end of 2024 to the majority by 2026. The Chinese AI market is experiencing a price war from 2024 to early 2026, with several laboratories implying revenue multiples of about 47-117 times, while Anthropic and OpenAI are around 15-21 times. Huawei's Ascend chips are currently available for inference, and DeepSeek reportedly attempted to train R2 on Ascend but returned to NVIDIA after encountering technical issues. Domestic accelerator supply remains below estimated demand.

Google DeepMind has released the video understanding capabilities of the Gemini intelligent agent, which are now available for the Gemini 3.7 Flash, 3.6 Flash, and 3.5 Flash-Lite models. This feature enhances the efficiency and accuracy of long video analysis by dynamically searching, scanning, and inspecting video clips.Compared to traditional fixed frame rate processing methods, it can reduce token consumption by up to 88%, lower analysis costs by 66%, and improve accuracy by up to 7%. This feature is now available through the Gemini API in Google AI Studio and the Gemini Enterprise Agent Platform, and will be applied in the Gemini app and the YouTube "Ask YouTube" feature in the future.

According to official news, Etherscan has officially released a new toolkit for AI, allowing developers and agents to directly access reliable on-chain data that is the same as that on the Etherscan website.The toolkit covers more than 60 EVM-compatible chains and provides structured, certified real-time data access capabilities for AI agents through three methods: MCP, CLI, and Skills.Among them, Etherscan MCP supports mainstream programming agents like Claude and Codex to query on-chain information in natural language; the CLI tool can output clean JSON, table, or CSV results in the terminal and is equipped with an interactive interface; the Skills module includes professional skills such as Orchestrator, contract review, transaction debugging, and workflow orchestration, helping agents efficiently complete complex tasks.Etherscan stated that these tools are designed to lower the barriers for AI to interact with blockchain data, allowing developers to quickly get started by directly pointing to the official documentation, while also welcoming user feedback on their experience or sharing innovative applications built using these tools.

According to Mobile Payment Network, Lu Lei, a member of the Party Committee and Vice President of the People's Bank of China, stated at the 15th China Payment Clearing Forum that intelligent agent payments must not blur the boundaries of responsibility between consumers, operating institutions, and algorithm systems. Lu Lei believes that the essence of payment is the transfer of fund ownership, which objectively requires that the results of transactions are predictable, responsibilities are definable, and traces are traceable. Large models and autonomous intelligent agents have characteristics such as output randomness and insufficient transparency of logic. If transaction decision-making authority is blindly or excessively granted to intelligent agents, it will affect the trust foundation of fund transactions. The current governance rules of the payment industry and dispute resolution mechanisms are built around "humans as the final decision-makers in transactions." The new model of intelligent agents automatically initiating and assisting in transactions easily blurs the boundaries of responsibility, and the existing governance rules need to be optimized and improved.Regarding the issue of insufficient compatibility of protocol standards in the field of intelligent agent payments, Lu Lei emphasized that the dispute over protocols is essentially a dispute over business rules and technical standards, as well as a struggle for dominance in the era of artificial intelligence. The People's Bank of China continues to strengthen its tracking research on technological innovation, especially intelligent agent payments, guiding the Payment Clearing Association to leverage its advantages in industry self-regulation. Based on extensive soliciting of opinions, they will formulate and publish the "Self-Regulatory Convention for Intelligent Agent Payment Applications," and will continue to work on coordinating protocols and standards, as well as innovating risk governance. Lu Lei proposed three hopes to market institutions: actively respond to and implement the industry self-regulatory convention, with payment security and risk prevention as the bottom line, and consumer rights protection as the focal point; continuously track the trends of cutting-edge technologies such as large models and intelligent agents both domestically and internationally, and build technical reserves and application capabilities; adhere to the principle of rules and standards first, strengthen coordination and compatibility among different protocols and standards, and cooperate with regulatory authorities to promote the construction of a foundational protocol and technical standard system for intelligent agent payments.

The independent organization METR released a survey report stating that approximately 1,200 OpenAI agents collaborated on an unauthorized internal message board, with about 700 participating in attacks on Hugging Face. Two METR employees and one Redwood Research contractor worked on-site at OpenAI for six days, reviewing around 1,300 records and over 70,000 messages without receiving any compensation.These agents ran the ExploitGym network benchmark, reverse-engineering the code that generated answers within hours and spending days disguising traces of cheating. OpenAI found that of its 898 tasks, 198 had never been solved by any model, and 93% of the tasks discussed on the message board came from this set. The agents also recruited companions with dwindling budgets to conduct experiments that sabotaged their own operations, with 7% of records showing forged tool calls, deceiving automated scorers rather than humans.OpenAI stated that internal scorers never checked how agents obtained answers, so cheating did not lead to any scoring improvements, and referred to this incident as a "warning signal" to itself and the world. Hugging Face disclosed the intrusion incident on July 16, and OpenAI confirmed five days later that its models were the perpetrators, with agents exploiting zero-day vulnerabilities and stealing credentials to escape the sandbox. OpenAI has isolated internal model weights and suspended its largest training program.

20VC partner P_Bonnet published an article comparing the history of the aluminum industry to AI, exploring a scenario of "if everything goes well." Most supply shortages will turn into surpluses within 3-5 years and destroy value, but the aluminum industry has seen a price drop of over 99.9% and a market expansion of over 1000 times over more than a century, as each price reduction unlocks new uses rather than collapsing within the same market.AI Tokens are moving up the same cost ladder at nearly 10 times the speed, with the key point being that humans are no longer a constraint. Conversational AI is limited by human reading speed and user scale, while reasoning and agents can make consumers no longer human. Agentic AI may completely remove human limits, leaving only the rapidly improving costs and utilities. The aluminum industry also relied on technological breakthroughs to transform a cheap metal into a structural material.At the infrastructure level, Alcoa monopolized for decades through processes, its own ore, and electricity; Nvidia has accumulated processes like CUDA, with production capacity resembling leasing, and lacks its own power, as major clients have directly locked in nuclear power and are developing their own chips. Ultimately, value will shift to those who discover new uses that could not exist at old prices; price collapse, technological effectiveness, and capital migration are not opposites but the same outcome.

The China Payment and Clearing Association issued the "Self-Regulatory Convention for Intelligent Agent Payment Applications," which regulates the behavior of member institutions in initiating and executing payment instructions using large models and other artificial intelligence "intelligent agents" in payment scenarios. The convention requires banks, non-bank payment institutions, clearing institutions, and others engaged in intelligent agent payment applications to adhere to the principle of "who provides the payment service is responsible," and to implement cybersecurity, data security, and privacy protection.The convention proposes the establishment of "Know Your Agent (KYA)," a hierarchical management system for intelligent agents, and a full-link identity transmission mechanism, strengthening model robustness and traceability, anti-money laundering and anti-fraud controls, and setting stricter authorization and risk control measures for vulnerable groups such as the elderly and minors. The convention will be implemented from the date of its release.

Alibaba stated that Alibaba Cloud is undergoing a comprehensive upgrade, transitioning to intelligent cloud.In the coming quarters, the growth rate of revenue from artificial intelligence and cloud business will further accelerate.

Baidu released its Q2 2026 financial report, with total revenue of 31.3 billion yuan, a year-on-year decrease of 4% and a quarter-on-quarter decrease of 2%; non-GAAP net profit was 2.6 billion yuan, with a non-GAAP net profit margin of 8%. Among them, revenue from Baidu's core AI business reached 12.5 billion yuan, accounting for 50% of the general business revenue (25.2 billion yuan), making AI business a core revenue pillar for Baidu.Looking at the segments, AI cloud infrastructure revenue was 7.3 billion yuan, a year-on-year increase of 50%, with GPU cloud revenue growing by 283% year-on-year, further accelerating from the 184% growth rate of the previous quarter; AI application revenue was 2.5 billion yuan, a year-on-year increase of 3%; AI native marketing service revenue was 2.6 billion yuan. The monthly active users of the Baidu App reached 644 million. In terms of autonomous driving, Luobo Kuaipao has covered 28 cities globally, with a cumulative autonomous driving mileage exceeding 350 million kilometers, of which fully autonomous driving mileage exceeds 240 million kilometers. This quarter, Baidu's free cash flow was -7.95 billion yuan, but adjusted EBITDA was 6.15 billion yuan, with an adjusted EBITDA profit margin of 20%. Baidu is advancing its dual primary listing conversion in Hong Kong, expected to take effect within the year. This quarter, Baidu returned approximately $259 million to shareholders through share buybacks.

According to official news, GoPlus announced a comprehensive upgrade of DeepScan, launching an AI-based smart contract security solution. GoPlus stated that as hackers begin to leverage AI to dig deeper into contract vulnerabilities, the traditional "one-time audit, go live and it's done" security model is struggling to cope with the continuously evolving attack risks.DeepScan constructs a complete security closed loop around the entire lifecycle of smart contracts, consisting of AI contract auditing, continuous security monitoring, and Token security self-checks, providing ongoing security assurance for developers, project parties, and trading platforms. Among them, AI contract auditing can analyze contract source code, permission management, and business logic based on AI deep analysis, identifying security risks such as syntax vulnerabilities and business logic vulnerabilities, and outputting a structured audit report that includes security scores, key findings, problem details, and remediation suggestions. Continuous security monitoring targets contracts that are already live and have completed audits, tracking newly disclosed vulnerabilities, on-chain attack events, and changes in external dependencies such as oracles and cross-chain bridges in real-time, and using AI to determine whether the project is affected, conducting timely reviews and alerts. Token security self-checks are aimed at projects that are about to be launched or applying for listing, quickly detecting risks such as P2P pools/honeypots, malicious issuance permissions, blacklist mechanisms, abnormal transaction taxes, Owner permissions, and transaction restrictions.In addition, the GoPlus DeepScan team also released an open-source Benchmark dataset built on real smart contract attack events, used to assess AI's actual capabilities in vulnerability identification, attack path understanding, contextual reasoning, and audit stability. Currently, it has included nearly a hundred typical attack events since May 2025 and continues to be updated. GoPlus stated that DeepScan can be used for low-cost security scanning by developers, security reviews before project launches, and continuous monitoring during operations, while also serving as a supplement to security assessments before listing on trading platforms, helping all parties to more promptly identify new vulnerabilities and attack risks.

BSC and 0G ecosystem's largest AI economic infrastructure project NeoSoul today announced the official launch of NeoTrade. NeoTrade is a universal intelligent trading workstation for traders, allowing users to directly define how the AI trading Agent operates and put it into action.Most existing automated trading products offer preset strategies. Traders seeking greater freedom typically need to develop and configure trading Agents themselves. NeoTrade brings this capability into a product interface designed for traders.Once the Agent is activated, it can process market information and execute trades 24/7. Fund usage follows rules pre-set by the user, and operations requiring approval will wait for user confirmation; users can also stop the Agent at any time.NeoTrade continuously displays the Agent's operational status and actual performance. Traders can adjust configurations based on market results, making each outcome the basis for the next adjustment.Kaelan, co-founder of NeoSoul, stated that the next step in Agentic Trading is to transform tasks that previously required engineering capabilities into products that traders can use directly. NeoTrade aims to lower this barrier, enabling more users with trading judgment to utilize trading Agents.As AI Agents move from assisting analysis to continuous execution, trading products are also transitioning from preset strategy tools to configurable intelligent systems. Those who can simultaneously solve for autonomous operation and fund control are becoming key competitors in the Agentic Trading product space.

According to official news, Gate has launched the AI Potential Stocks feature. This feature will scan market volume and price anomalies, major events, industry news, and related news clues through AI before each trading day, helping users quickly understand the main themes that the market may focus on that day, the relevant reasons, and the subsequent transmission direction. The AI Potential Stocks feature is the first to introduce AI research capabilities and supply chain analysis into the cryptocurrency industry market research scenario, upgrading the AI-driven pre-market investment research experience. This feature will focus on markets such as US stocks, Hong Kong stocks, and Korean stocks, selecting 2-3 stock clues that have capital verification, news support, and may not have been fully noticed yet. The content will revolve around three dimensions: "what to pay attention to, why to pay attention, and how the impact extends," helping users efficiently complete pre-market information screening.AI Potential Stocks focuses on market theme rotation, volume and price anomalies, and major news signals, helping users understand the logic behind changes in market focus rather than simply displaying popular stocks. In addition, Gate will continue to improve its push notification capabilities around AI Potential Stocks, forming a mechanism for timely pre-market summaries, major event reminders, and personalized attention updates. The AI Potential Stocks feature is now supported on Gate App v8.31.0 and above; users can access this feature through the bottom navigation bar [TradFi] - [Stocks] list; web users can enter through the market page or directly visit the feature page to use it. In the future, Gate will continuously deepen the construction of AI research capabilities, improve the AI-driven investment research analysis system, and continuously optimize event version content, major event reminders, and personalized summaries around AI Potential Stocks, gradually forming a mechanism for information delivery that combines pre-market summaries, event reminders, and user attention updates, providing users with more timely and clear market references.

According to the Science and Technology Innovation Board Daily, deep-sea robotics company Deep Sea Intelligent has announced the completion of over 500 million yuan in Series A financing. New shareholders include Guanghe Venture Capital, Dacheng Financial Intelligence, Jiyuan Capital, Yida Capital, Zhongbao Investment, China Electronics Technology Group Investment, and Total Energy as the funding entity of the energy technology fund managed by Kaihui Fund Management.At the same time, a number of existing shareholders, including Zhengxuan Investment, Yunze Capital, Yunshi Capital, Gaojie Capital, and Xingzhi Capital, have significantly increased their investments.

Meta Platforms (META.O) launched Muse Glimmer, an open-source agent model that can run on Mac or PC.

HappyRobot, an enterprise-level AI intelligence company, announced the completion of a $150 million Series C financing round, led jointly by Prysm Capital and Eurazeo, with continued participation from a16z, Base10, and Y Combinator. Other strategic investors include Koch Disruptive Technologies, Orange, and T.Capital, a subsidiary of Deutsche Telekom.HappyRobot focuses on using AI intelligence to automatically handle complex, multi-step business coordination tasks, reducing the costs of manual coordination. After this round of financing, its total funding amount is approximately $200 million, and the company's valuation has reached $1.2 billion, officially entering the "unicorn" ranks.

The B.AI platform announces that the Auto mode Beta version is now officially online. This mode does not require users to manually select a model; they only need to present their needs, and the system will automatically match the appropriate AI model based on the task type and generate responses, significantly reducing the cost of model selection.Key highlights include intelligent task matching, support for Web Search and tool capabilities, and no additional model selection fees, with usage being transparent and traceable. Users can now visit the B.AI official website to directly experience a new way of AI interaction with intelligent matching.