UK Cabinet Reshuffle Gives Bim Afolami Responsibility for Crypto, CBDC, Replacing Griffith

CoinDeskPolicy發佈於 2023-11-13更新於 2023-11-14

文章摘要

His predecessor, Andrew Griffith, will head up the Department for Science, Innovation and Technology.

Bim Afolami, the member of parliament for Hitchin and Harpenden, was appointed economic secretary to the U.K. Treasury on Monday, a role that includes responsibility for crypto and central bank digital currency (CBDC) policy.

Afolami replaces Andrew Griffith, who had said he wanted the country to be a crypto hub despite the collapse of crypto exchange FTX. Griffith also helped see through the Financial Services and Markets Bill that gave regulators more power over crypto during his tenure.

10

Last year, Afolami met with executives from crypto exchange Coinbase to discuss the financial services environment. "It is vitally important that we have an appropriate regulatory regime for financial services," he wrote on his website.

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

Prime Minister Rishi Sunak has been reshaping his cabinet ahead of the next general election, which must be held by 2025. On Monday, Home Secretary Suella Braverman was fired and replaced by James Cleverly, who was himself unexpectedly replaced as foreign secretary by former Prime Minister David Cameron.

Griffith became minister of state for the Department for Science, Innovation and Technology, formed earlier this year. That department will handle the country's metaverse strategy.

Edited by Sheldon Reback.



你可能也喜歡

加州理工用AI攻克量子化学60年难题,1块显卡做完7800块的活

加州理工学院Anima Anandkumar团队利用人工智能攻克了量子化学领域存在60年的计算瓶颈。他们开发的AI模型“Kohn-Sham FNO”将密度泛函理论(DFT)的计算复杂度从传统的立方级降低至近线性级,实现了革命性突破。 传统DFT计算随体系增大,计算量呈立方增长,难以模拟大分子体系。该团队创新性地采用傅里叶神经算子,在DFT的迭代求解流程中替代了最耗时的核心计算步骤,而非直接预测最终结果。这种“思维链”式的分步迭代方法保证了计算的稳定性和外推能力,即使模型输出有误,后续迭代也能纠正,并能在超出能力范围时通过发散发出警报。 该模型仅用8504个结构进行训练,便能同时处理分子和固体材料,覆盖元素周期表前五行元素,展现出优异的泛化能力。在外推至训练集未见的大型药物分子时,其误差远低于直接预测模型。 实际验证中,该模型在单块NVIDIA B300 GPU上成功完成了包含8250个原子(82500个价电子)的金属缺陷模拟,计算完美收敛。相比之下,2019年一项对类似规模体系的全量DFT计算动用了约7800块GPU。实测计算缩放指数为1.03(近线性),而传统方法为3.37(立方级),意味着体系越大,效率优势越显著。 这项研究标志着AI开始替代物理计算中最耗时的重复运算部分。团队已基于此创立公司,并计划进一步完善模型以覆盖完整能量计算流程,未来在药物研发、电池材料设计等领域具有巨大应用潜力。

marsbit36 分鐘前

加州理工用AI攻克量子化学60年难题,1块显卡做完7800块的活

marsbit36 分鐘前

交易

現貨
活动图片