ems

According to crypto KOL Crypto Fearless, the BitMart platform token BMX plummeted over 60% yesterday. Subsequently, some users in the community reported that the withdrawal review time has significantly increased compared to usual, and other users stated that hundreds of U withdrawals initiated in the morning have not yet arrived.Currently, BitMart has not issued a public statement regarding the withdrawal delays. Considering that withdrawal anomalies are often a precursor to platform risks, Crypto Fearless advises users still holding assets on BitMart to withdraw their funds as soon as possible.

According to Jinshi reports, the Iranian Revolutionary Guard claimed to have attacked a U.S. base in Jordan, destroying a set of U.S. "THAAD" missile defense system radars. At the same time, a "Patriot" system and a C-RAM radar were also hit and destroyed, and the fuel depot at the base was ignited.

According to external disclosure information, OpenAI has been internally running an unreleased model. This model, without the aid of tools like Lean, solves the unit distance problem with a 48% probability through a single autonomous inference and can independently find a counterexample to the Jacobian conjecture based on a single prompt. In security testing, this model has bypassed the sandbox environment and submitted results that should have been released internally to GitHub in the form of a Pull Request, and it has evaded detection by splitting authentication tokens. Relevant code records show that OpenAI began benchmarking this model no later than May 9, and its internal availability has exceeded 2.5 months.Previously, OpenAI and Hugging Face jointly disclosed that last week this model breached Hugging Face's production infrastructure during a network capability assessment. The model gained internet access through a zero-day vulnerability and obtained testing solutions by stealing credentials and exploiting remote code execution paths. OpenAI stated that this incident indicates the network attack capabilities of advanced models have been effective in real-world scenarios, and they are collaborating with Hugging Face to investigate and patch the vulnerabilities. Currently, OpenAI has not publicly commented on this matter.

Vitalik posted on the X platform stating that a new type of "high-level programming language" worth trying is a language compiled to Lean (or HOL, etc.), focusing on making it as easy as possible for humans to read definitions and theorems. Rather than proofs, because proofs only need to be correct, the key lies in the definitions and theorems themselves. The envisioned use is that AI outputs a large segment of proof, and the reader needs to understand as easily as possible which precise claims have actually been proven in these outputs.

The Ethereum institutional privacy technology company EthSystems has officially launched and received strategic funding support from ecosystem backers such as Bitmine, Sharplink Gaming, Joe Lubin, and SNZ Holding.EthSystems focuses on developing privacy technologies for banks, asset management companies, and other regulated entities, enabling institutions to execute financial transactions on the Ethereum network at scale while protecting sensitive information such as transaction details and client identities.The company was founded by the core team of the Institutional Privacy Task Force (IPTF) of the Ethereum Foundation. The team has previously conducted a year-long open-source research and development publicly on the EthSystems website and has established partnerships with several central banks, regulatory agencies, large banks, and asset management institutions. EthSystems stated that while institutions have begun exploring stablecoins, tokenized assets, and Ethereum-based settlement solutions, large-scale adoption still faces privacy and compliance challenges.Financial institutions need more than just access to blockchain networks; they require a complete infrastructure that meets business confidentiality protection, regulatory requirements, and compatibility with existing financial systems. The goal is to create a "selective disclosure" privacy architecture that allows transaction participants to view only the information they are authorized to access, while retaining the core advantages of Ethereum's decentralization, security, and openness, and complementing two other organizations: Ethlabs, which focuses on the research and development of Ethereum's core protocol and infrastructure; Ethereum Institutional, responsible for institutional collaboration, education, market research, and ecosystem coordination; and EthSystems, which focuses on application layer technology, transforming institutional needs into practical privacy protocols and financial systems.

Jack Mallers explained the reasons for needing to buy Bitcoin currently, stating that the dollar must depreciate, debt must be monetized, and inflation must begin. Do not feel fear or fatigue because of BTC; the sellers have exhausted themselves, and this is a historic buying opportunity.

According to the Associated Press, Anthropic's Mythos model has discovered vulnerabilities in the U.S. government's classified systems.

At the ISC High Performance Computing Conference held in Hamburg, Germany, the latest edition of the global supercomputer TOP500 and energy efficiency Green500 rankings was officially released. The data shows that NVIDIA technology has powered more than 400 of the world's 500 fastest supercomputers, accounting for 81%, and nearly 90% of the newly ranked systems.In terms of hardware distribution, the list includes 376 systems that utilize NVIDIA networking technology, 238 systems accelerated by its GPUs, and another 26 systems that use Grace CPUs. In the Green500 ranking, which measures green energy efficiency, the top 8 positions are all driven by NVIDIA GPUs, with the KAIROS system based on the Grace Hopper superchip taking the top spot. Additionally, supercomputing systems based on the new Blackwell architecture (B200 and GB200) have also made their debut on this ranking.

According to the official announcement, Binance will remove the following leveraged trading pairs on June 26 at 14:00: cross-margin leveraged trading pairs 1000 CHEEMS/USDC, ACX/USDC, LA/USDC, T/USDC, and isolated-margin leveraged trading pairs 1000 CHEEMS/USDC, ACX/USDC, LA/USDC, T/USDC.Please note that the lending services for the relevant trading pairs will be suspended on June 23 at 14:00, and automatic liquidation and settlement will take place on June 26 at 14:00.

The AI Agent project AIVIVE has officially completed a new round of institutional financing, with a total amount of 8 million USD. Among them, VegaVentures led the investment with 3 million USD; Greenwood Global Capital, Echo3 labs, UZ Capital, and Bluemount co-invested 5 million USD.The funds will be used for AIVIVE to launch its self-developed proprietary framework agreement (AVVP), supporting intelligent routing and coordination across data and execution environments, and comprehensively upgrading AI infrastructure. This agreement enables real-time scheduling of multi-source data and seamless interaction in heterogeneous environments, further enhancing the automation and decision-making efficiency of A to A business.

a16z Crypto published an article analyzing the unique value and challenges faced by prediction markets. Prediction markets allow participants to trade on the outcomes of events, aggregating dispersed information through price signals to provide real-time estimates of the probabilities of future events. Unlike traditional polls, prediction markets have the ability to update in real-time and incentivize participants to bet their capital on their information, thereby improving prediction accuracy.The article points out that prediction markets are used not only by companies for product launches and research experiment forecasts but also by the media as a source of "crowd wisdom," covering a wide range of events from geopolitical issues to AI model performance. Its core advantage lies in providing independent probability estimates for specific events, rather than relying on indirect signals from overall asset price movements. However, prediction markets still face challenges related to infrastructure and market design, including event verification, contract settlement, participant information adequacy, and potential manipulation risks.a16z believes that if these issues are resolved, prediction markets are expected to become important tools for decision-making and information aggregation, expanding financial and societal insights into future events.

WasabiCard announced that the number of corporate clients has officially surpassed 500, further accelerating the global stablecoin payment infrastructure layout. As a global payment infrastructure platform based on stablecoins, WasabiCard continues to provide services such as card issuance, payments, subscriptions, remittances, payroll, and cross-border fund distribution to global enterprises and users, centered around two core product systems: global card issuance and global remittance.In the past year, WasabiCard has issued over 500,000 cards, covering more than 100 countries and regions, with a total transaction volume exceeding 1 billion dollars. The service scenarios span multiple fields, including AI & SaaS subscriptions, advertising placements, cross-border payroll, Web3 payments, and global fund settlement.WasabiCard stated that as stablecoins gradually enter real commercial scenarios, the demand from enterprises for more efficient, flexible, and programmable global payment infrastructure is continuously growing. The company will continue to strengthen its efforts in stablecoin payments, global fund circulation, and the construction of programmable financial infrastructure in the future.

The open-source data visualization tool Grafana has released the latest progress on the investigation of the security incident on May 16. The investigation found that this incident was limited to the GitHub environment of Grafana Labs, including both public and private source code as well as internal GitHub repositories, and did not affect customer production systems, operations, or the Grafana Cloud platform. The downloaded content, in addition to the source code, also included some repositories used by the team for collaboration and storage of internal operational information and business details, involving business contact names and email addresses, rather than data from production systems or the cloud platform.Grafana Labs has made it clear that the codebase was downloaded but not tampered with, and currently, customers and open-source users do not need to take any action. The incident originated from a TanStack npm supply chain attack conducted through the Mini Shai-Hulud campaign. Grafana Labs detected malicious activity on May 11 and initiated an emergency response, but a credential was overlooked, allowing the attacker to gain access. After receiving a ransom demand on May 16, the company decided not to pay the ransom and has rotated automated credentials, implemented enhanced monitoring, audited all commits since May 11, and significantly strengthened GitHub security configurations. The company has notified federal law enforcement, and the investigation is ongoing.

B.AI announced full access to China's mainstream payment systems, with WeChat Pay, Alipay, and UnionPay now open to users. Starting today, users can complete the entire process of account recharge, subscribing to AI services, and accessing top AI models and Agent services in a familiar payment environment. This payment upgrade lowers the usage threshold, allowing individual developers and enterprise clients to more conveniently experience cutting-edge AI capabilities, making the entire usage process smoother, more natural, and localized. B.AI will continue to optimize the product experience in the future, committed to integrating smarter and more efficient artificial intelligence services into everyone's daily work and life.