Japan’s Monex Group Eyes Yen-Backed Stablecoin Launch

TheCryptoTimes發佈於 2025-08-27更新於 2025-08-27

Tokyo-based financial company Monex Group is thinking about launching a stablecoin, which will be pegged to the Japanese yen.

According to TV Tokyo, Chairman Oki Matsumoto said that issuing a stablecoin requires significant planning and investment, but it is necessary to stay competitive in digital finance. He added, “We will respond appropriately.”

The stablecoin would be backed by assets like Japanese government bonds and could be exchanged 1:1 for yen. It could be used for things like international money transfers and business payments. 

Monex plans to use its local crypto exchange, Coincheck, and its securities business to grow this initiative. The company is also considering buying European crypto companies, with an announcement expected soon. 

Japan Embraces Stablecoins

Japan’s Financial Services Agency (FSA) is preparing to allow Japanese yen stablecoins later this year, which would be the first domestic fiat-pegged stablecoin in the country. Earlier, Japan also approved the use of the US dollar-backed stablecoin USDC in March. These changes follow Japan lifting its ban on foreign stablecoins in 2023.

Stablecoins are also gaining attention worldwide. In the US, new laws have strengthened dollar-backed stablecoins, while in Japan, companies like JPYC and SBI Holdings are also entering the market.

SBI Holdings recently teamed with Ripple to introduce Ripple USD (RLUSD), a USD-backed stablecoin, in early 2026. The company has also collaborated with Circle to promote USDC and create digital finance services in Japan.

Also Read: Crypto Can Be a Part of Diversified Portfolios: Japan Finance Minister



你可能也喜歡

GPT5.6惨遭切脑,Fable 5回归要变弱鸡版?

近日,AI领域两大巨头OpenAI和Anthropic的最新顶级模型遭遇重大监管介入。OpenAI刚刚推出的GPT-5.6系列被拆分为Sol(前沿模型)、Terra(均衡模型)和Luna(经济模型)三个版本。其中性能最强的Sol版本并未全面发布,其访问权限被限制在“获批名单”内,绝大多数用户需等待数周才能逐步使用。此举源于监管机构在全美AI安全框架落地前,对具备“类似Mythos”能力的尖端模型进行预先管控。 在此之前,Anthropic的Fable 5模型经历了更剧烈的风波。该模型发布仅72小时后即被全球禁用,原因是其底层技术Mythos在一次闭门演示中展示了强大的网络攻击能力,例如能找出银行系统漏洞并清空账户。经过安全审查,Mythos 5仅向极少数“信任伙伴”有限开放。据报道,即将回归的Fable 5将是“阉割版”,安全护栏被大幅加强,可能导致能力下降和用户体验受限。 行业观察指出,严格的监管可能迫使AI公司对模型进行“脑叶切除”般的深度限制,以通过安全审查。这引发了开发者对“货不对板”的担忧:用户最终获得的可能是一个能力被严重削弱、反应僵化的“傀儡”,而非当初期待的“天才”。同时,模型的获取方式也可能从“付费即用”变为需要严格身份验证或额外收费。这一系列事件标志着AI发展进入“算法配给制”时代,引发对创新与安全平衡的广泛讨论。

marsbit10 分鐘前

GPT5.6惨遭切脑,Fable 5回归要变弱鸡版?

marsbit10 分鐘前

老黄:Prompt已死,整个AI圈都在疯狂追Loop

近期,硅谷AI圈掀起“Loop Engineering”(循环工程)热潮,业界普遍认为传统手动编写提示词(Prompt)的模式即将被淘汰。英伟达CEO黄仁勋、吴恩达、Anthropic及OpenAI等多位专家和机构均指出,未来核心不再是设计单次指令,而是构建能够自主运行、迭代的AI循环系统。 循环工程的核心在于将人类从实时干预中解放出来,转变为系统架构师。人类只需一次性定义目标、停止条件、验证机制、记忆存储及调度规则,随后AI系统便可自动执行“发现任务→执行→验证→持久化→再发现”的闭环流程,实现24/7不间断工作。其中,独立的验证环节尤为关键,需设立专门的评估智能体对输出进行客观审查,避免自我美化。 这一转变意味着代码生成成本大幅降低,而人类的判断力成为关键稀缺资源。然而,全自动循环也可能带来验证债务、代码理解脱节、认知依赖及token消耗失控等风险。因此,成功的循环设计需融入严谨的判断与监督,否则可能放大错误或导致技术债务。 目前,相关实践指南和白皮书已在网络广泛传播,标志着AI开发正从“提示工程”迈向“循环工程”的新范式。尽管自动化能力增强,但工程师的核心角色并未消失,而是升级为循环系统的设计者与决策者,确保AI在正确方向上高效运行。

marsbit10 分鐘前

老黄:Prompt已死,整个AI圈都在疯狂追Loop

marsbit10 分鐘前

交易

現貨
活动图片