Odaily News: On October 8, Metis announced the launch of the Metis Rise monthly developer incentive program, rewarding teams that deploy new AI Agents on the Metis Andromeda mainnet, acquire real users, and generate real on-chain economic activity.Metis stated that while many teams can currently build Agents capable of answering questions, Agents that can get paid, prove their identity, and earn sustained transaction trust remain scarce. Metis Rise aims to support developers in exploring this direction, help developers innovate, and drive the growth of AI Agent applications within the Metis ecosystem.Metis also stated that it is building an execution and data layer for AI Agents, and that Andromeda's decentralized sequencer (DSeq) can eliminate the single point of failure that trustless transactions between Agents depend on.The program focuses on the "economic usability" of AI Agents, covering four categories of needs: payments (x402), identity (ERC-8004), stablecoin escrow (Escrow), and independently verifiable on-chain signature receipts.In terms of participation thresholds, developers need to meet the following conditions each month:- At least 15 valid monthly active users;- Each counted user must complete at least 3 qualifying transactions on Metis Andromeda, or settle an amount of no less than $2 (m.USDC, m.USDT, or METIS);- Primary user transactions, settlement volume, and Gas consumption must occur on the Metis Andromeda mainnet.The program is open to all developers, but is limited to projects newly deployed on the Andromeda mainnet. It supports any AI Agent framework and hosting method, including ClawUp, GOAT AgentKit, or self-built backends, and supports a dual-chain architecture based on GOAT Network. Metis Rise is hosted by Metis, with ClawUp, GOAT Network, and LazAI as ecosystem partners.

Odaily News: Vlad Tenev posted on X that Robinhood Chain is a top priority for the company, the company has never left, and developers who participated in building it early on are important to the company. Robinhood is dedicating resources to attract excellent developers to build on Robinhood Chain, and will announce more details soon.

Odaily News: Anthropic's transparency report updated on July 23 shows that in the first half of 2026, a total of 11.4 million accounts were banned, an increase of about 6.9 times compared with 1.45 million in the second half of 2025, and the total number of bans was about 7.9 times that of the previous period.During the same period, Anthropic received a total of 398,000 ban appeals, of which about 42,000 overturned the original ban decisions, with an appeal success rate of about 10.6%.Anthropic stated that reasons for account bans include violations of usage policies, terms of service, or supported regions policy. The relevant statistics cover all of its products and services and do not separately distinguish the Claude chat product from developer services.

Odaily News: The XRP Ledger Foundation posted on the X platform to clarify the recent XRP Ledger security vulnerability incident, stating that the vulnerability could potentially have been used to bypass the limit that fixes XRP's total supply at 100 billion, but it was fixed two weeks ago via version 3.4.1. Furthermore, upon verification, no new XRP was found to have been created, and the total amount of XRP currently circulating on the ledger is approximately 99.986 billion.The Foundation stated that the vulnerability would have required hundreds of deliberately constructed operations to trigger, and normal payment transactions would almost never involve it. In the future, it will strengthen formal verification, AI adversarial testing, security research collaboration, and bug bounty programs with RippleX and the community. Regarding the fix process, the Foundation explained that publicly releasing a patch could allow attackers to use AI to accelerate vulnerability analysis. To address this situation, the XRPL community engaged in coordination. Participating parties including the XRP Ledger Foundation, RippleX, core developers, and validators reached a social consensus and agreed to adopt the approach of directly releasing the fix. In addition, version 3.4.1 only became a version the network was required to adopt after going through the normal amendment voting process and receiving support from more than 80% of validators for two consecutive weeks. Protocol changes still need to be completed through the amendment mechanism.Previously, RippleX engineering lead J. Ayo Akinyele also posted on the X platform, stating that he attaches great importance to the recently disclosed critical vulnerability in the XRP Ledger (XRPL) payment engine. This vulnerability is a serious security issue that has existed in the codebase for more than 10 years and could theoretically lead to the over-creation of XRP, breaking through the fixed supply cap. However, before the vulnerability could be exploited, it was discovered by the security research firm Veria AI through Ripple's bug bounty program.

According to Odaily, Bitcoin News posted on X that the EU is reviewing MiCA, and the related consultation involves requiring developers of non-custodial Bitcoin wallets to obtain certification before offering software to the public. The European Securities and Markets Authority (ESMA) stated that open-source development, self-custody, automated smart contracts, and permissionless infrastructure should not automatically be regarded as regulated intermediaries, while also proposing a new regulated service for crypto asset service providers that offer clients access to decentralized protocols. This review has not yet changed the current rules. The European Commission will evaluate the relevant feedback and, in accordance with Article 140 of MiCA, submit a review report by June 30, 2027.

Odaily News: Microsoft CEO Satya Nadella published a long article stating that as Super Intelligence systems develop, the industry needs to re-examine AI trust architecture, and cannot simply rely on commitments from model providers, nor outsource the responsibility for AI behavior.Nadella said that over the past few decades, when traditional software systems were deployed across industries, developers could usually trace behavior to specific code paths. But today's frontier AI models have capabilities beyond traditional software systems, yet cannot clearly attribute model outputs to specific training data or model weight configurations. Enterprises are connecting AI systems with autonomous action capabilities to sensitive data and granting them permissions to perform critical tasks, so they need to build systems with observability, testability, and controllability, separating "intelligence supply" from "permission control."Nadella proposed that superintelligence should be governed through an engineering approach, constraining non-deterministic models through deterministic system design, human control, and operational processes, and managing AI systems as potential "insider risks," including limiting permissions, recording behavior, and establishing isolation boundaries. He also proposed that superintelligence systems should follow principles such as multi-model collaboration, end-to-end observability, continuous verification, independent control, independent auditing, risk isolation, and incident disclosure. Among them, he emphasized that model chain-of-thought (CoT) transparency is a baseline requirement, but CoT transparency alone is not enough, because model outputs themselves may still lack reliable interpretability.Nadella concluded by saying that the most trustworthy superintelligence system in the future is not the "most trustworthy model," but a "system that can still operate safely even if the model is not fully trusted."

According to GMGN market data, the market cap of SOL ecosystem Meme coin CLAUDIA briefly surpassed $9 million, before falling back to approximately $8 million, with an intraday gain of over 580%.Regarding the related narrative, CLAUDIA is built around an AI character IP designed by AI creator Anabology, and has recently drawn attention from a16z co-founder Marc Andreessen. Community opinion holds that CLAUDIA's core narrative lies in its AI character IP and its commercialization potential. The developer recently interacted with Musk and plans to use trading fees for AI film production, with related merchandise potentially launching in the future.Odaily reminds readers that Meme coin prices are highly volatile and susceptible to market sentiment and trending events. Please be aware of trading risks.

Odaily News: NEAR treasury company SVRN has announced the acquisition of NEAR infrastructure platform FastNEAR. FastNEAR will be incorporated into SVRN as a wholly-owned subsidiary, and its co-founders Evgeny (Eugene) Kuzyakov and Mike Purvis will also join the SVRN team. FastNEAR is a high-performance RPC infrastructure provider for NEAR applications, supporting the majority of NEAR's data layer, including server clusters that handle network read and write operations, archival infrastructure that stores complete transaction history records, and NEARDATA, a data source for developers to process historical records.

Odaily: Arc has announced that MoonPay has officially integrated with its blockchain network. MoonPay, which has over 30 million users, supports purchasing USDC via bank cards, Apple Pay, Google Pay, and bank transfers, with funds transferable to Arc. This integration will provide fiat on-ramp channels for wallets, applications, and developers within the Arc ecosystem.

Odaily News: AI startup TypeSafe AI has completed an approximately $870 million funding round led by venture capital firm Andreessen Horowitz (a16z), bringing the company's valuation to $7.5 billion.TypeSafe AI's AI model Jev quickly gained traction within weeks of its release. Co-founder Diogo Almeida said that after the launch video went viral, the company attracted significant attention from corporate executives, developers, and investors. Jev is currently used by approximately one-third of Fortune 500 companies and surpassed 1 million users within days. (Bloomberg)