U.S. Banking Watchdog Hsu Says Tokenization Promising, But Crypto Full of Fraud

CoinDeskPolicy發佈於 2023-11-06更新於 2023-11-07

文章摘要

Michael Hsu, the acting chief of the U.S Office of the Comptroller of the Currency that oversees banks, is excited about the possibilities of tokenization to solve settlement...

Tokenization of assets could be the answer to the risky complexities of settling the movement of funds and securities, said Michael Hsu, the acting head of the U.S. Office of the Comptroller of the Currency.

"Tokenization is focused on solving an actual problem, and that problem is settlement." Hsu said at the DC Fintech Week event in Washington. "This is boring back-office stuff, but it's super, super important."

10

Any time an asset changes hands in the financial world, the transaction typically passes through multiple entities and checks of its validity before it's cleared and settled to officially land in the hands of the recipient. At any of those layers – most of which carry their own expenses that may be added to what the customer pays – the transaction has some risk of failure.

A D V E R T I S E M E N T
A D V E R T I S E M E N T

"Tokenization holds the promise to collapse that and to simplify it, if it's done right," he said.

His OCC is so engaged on the idea of the tokenization of real-world financial assets and liabilities that it's hosting a Feb. 8, 2024, all-day discussion on the topic at its Washington headquarters. But when it comes to the rest of the crypto space, Hsu remains suspicious.

"There seems to be more and more of a divide between crypto on one hand and tokenization," he said. Crypto, he said, "tends to be driven by the hope for speculative gain.""It still remains replete with frauds, scams and hacks," Hsu said.

Edited by Nikhilesh De.

你可能也喜歡

造ChatGPT的人,已经不用ChatGPT干活了

不到一年时间,OpenAI已将内部主力AI工具从ChatGPT聊天机器人转向了智能体Codex。截至2026年6月,Codex处理了公司每周99.8%的输出token,而10个月前这一比例还不足10%。 这一转变始于去年9月左右,Codex在接入了更强模型、补齐更多能力后,能够承担更复杂的任务。员工发现,与其进行一问一答的对话,不如将整件长周期任务直接交给它自主运行。如今,OpenAI人均超过85%的输出token由Codex产生,各部门均将其作为头号AI工具。 智能体正在改变知识工作的基本单位:从单次问答转变为可“丢出去”独立执行数分钟到数小时的任务。如今近四分之一的Codex请求对应的是人类需耗时一小时以上的工作。 最初是工程师广泛采用Codex,但火势迅速蔓延至法务、财务、招聘等非技术部门。到2026年4月前后,这些部门集体转向,使用速度甚至超过工程部。一个关键信号是,大量原本不写代码的用户开始使用Codex。例如,财务团队用它处理了数万份税表,公关团队搭建了自动处理邀约的Slack智能体。业务人员用Codex完成的工作中,超过四分之一涉及编程,岗位界限正在模糊。 Codex的角色已从代码补全工具转变为通用工作流智能体,能独立完成包括实现、调试、测试在内的完整工程链。重度用户单日可调度Codex产出超过60小时的智能体工作量。其底层模型GPT-5.5不仅能高效处理长任务,甚至被用于优化自身的系统负载算法,提升了性能。 报告显示,办公的默认动作正从打开聊天框提问,转变为将整件事交给智能体执行。未来的差距将取决于人们敢于将多大规模的任务交由AI独立完成。

marsbit1 小時前

造ChatGPT的人,已经不用ChatGPT干活了

marsbit1 小時前

交易

現貨
活动图片