HSBC будет открывать токенезированные депозиты для пользователей из США и ОАЭ

cryptonews.ruОпубліковано о 2025-11-18Востаннє оновлено о 2025-11-19

HSBC расширяет деятельность в области цифровых технологий и планирует вывести токенизированные депозиты на новые международные рынки. По данным Bloomberg, банк намерен предложить профильную услугу корпоративным клиентам в США и ОАЭ в первой половине 2026 года. Это решение отражает растущий интерес мировых финансовых компаний к блокчейн-технологиями, которые способны существенно ускорить платежные процессы и повысить прозрачность операций.

Руководитель платежного подразделения HSBC Маниш Коли отметил, что токенизированные депозиты позволят проводить мгновенные транзакции в любое время, включая международные переводы. По его словам, компании стремятся модернизировать управление денежными потоками, что побуждает банки искать более гибкие решения. Услуга уже запущена в Гонконге, где первым клиентом стала организация Ant International. Сейчас же идет процесс запуска профильной услуги в Сингапуре, Великобритании и Люксембурге.

Токенизированные депозиты — это цифровая форма банковских вкладов, размещенная в блокчейне и полностью обеспеченная резервами самого банка. Такой подход делает расчеты мгновенными и открывает возможности для внедрения программируемых механизмов платежей. В отличие от стейблкоинов, выпускаемых небанковскими структурами, депозитные токены остаются в поле традиционного регулирования и позволяют клиентам рассчитывать на стабильный процентный доход, что особенно важно для корпоративного сектора.

HSBC также рассматривает возможность использования токенизированных депозитов в автономных казначейских системах. Они базируются на алгоритмах и элементах искусственного интеллекта для более эффективного управления ликвидностью.

Такие решения позволяют сократить время обработки финансовых операций и повысить точность бизнес-планирования, что становится все более актуальным для крупных международных компаний.

Пов'язані матеріали

Sam Altman's ChatGPT Parenting Method Stirs Controversy

Sam Altman, CEO of OpenAI, recently faced significant backlash after sharing a concept for using ChatGPT to manage family life. He suggested feeding the AI family schedules and children's interests to generate a personalized morning podcast for car rides, covering topics from soccer games to birthdays alongside news for parents. The post, intended to showcase a "cool" AI application, was widely criticized online. Critics, including "Gravity Falls" creator Alex Hirsch, argued the idea undermined genuine parent-child interaction, with Hirsch's rebuttal—"You could just... talk to your kids?"—gaining far more engagement than Altman's original post. This incident highlights a perceived contradiction in Altman's persona, who has previously expressed the pain of missing time with his child while now proposing AI-mediated conversations. This is not Altman's first promotion of AI-assisted parenting; he has previously described using ChatGPT as a 24/7 parenting encyclopedia for newborn care questions. However, the latest suggestion—having AI script family conversations—crossed a line for many, shifting his image from an anxious new dad to primarily an OpenAI promoter. The controversy coincides with OpenAI's strategic push into the "family AI" space, as evidenced by hiring a family product manager and shifting user demographics toward older parents. While AI can efficiently manage schedules and information, the strong public reaction underscores the sensitivity of applying AI to core human relationships. Critics emphasize that AI should assist with logistical tasks to free up time for family, not replace the empathy, spontaneous connection, and shared experiences that define genuine caregiving. The debate points to crucial boundaries for AI in the home, where efficiency is less important than authentic human connection.

marsbit3 хв тому

Sam Altman's ChatGPT Parenting Method Stirs Controversy

marsbit3 хв тому

Just Now, OpenAI's Largest Pre-trained Model Doug Exposed

On August 9, X user ChrisGPT reported that OpenAI is advancing a new large-scale pre-training model codenamed **Doug**, which is said to be its largest such project to date and distinct from GPT-6. ChrisGPT later suggested GPT-6 is likely Astra, a model OpenAI recently paused due to safety concerns, with Doug potentially launching by November. This aligns with a July 9 research memo from SemiAnalysis, which stated OpenAI has overcome pre-training issues and is actively developing a much larger model codenamed Doug. If accurate, this signals a potential shift: after nearly two years of relying primarily on post-training, reinforcement learning (RL), and inference-time compute for capability gains—evidenced by models like o1, o3, and the GPT-5 system—OpenAI may be restarting large-scale foundational model scaling. The backdrop includes competitive pressure from Google's Gemini 3 release in November 2025, which reportedly prompted internal focus at OpenAI. In December 2025, The Information reported OpenAI was developing a pre-training model codenamed **Garlic**, which showed promising results and incorporated key bug fixes. This project reportedly paved the way for an "even bigger and better model"—likely Doug. In summary, Doug may represent OpenAI's return to significant base model scaling, building on resolved pre-training challenges and aiming to push capabilities beyond the limits of the current GPT-4o-era foundation enhanced by advanced post-training techniques.

marsbit24 хв тому

Just Now, OpenAI's Largest Pre-trained Model Doug Exposed

marsbit24 хв тому

OpenAI Researcher: We Don't Read Papers Anymore

An OpenAI researcher's remark that top AI labs "no longer read papers" sparked widespread discussion, highlighting a deepening crisis of trust in academic publishing. This sentiment followed exposure of questionable practices in an ICLR paper, where exceptional results were linked to undisclosed "tricks." A large-scale "experimental review" by SAI of 168 Oral papers from ICML 2026 revealed severe reproducibility issues. Of the 105 papers fully replicated, only 8 successfully verified over 80% of their claims, with a median verification rate of just 28-30%. Common problems included non-runnable code, missing files, incomplete documentation, and results mismatching those reported. Some papers even relied on now-offline models, making verification impossible. Specific cases involved an 8x inflation in claimed trained parameters and missing evaluation models from released code. Verifying a single ICML Oral paper had a median cost of around $8,900, with 17 exceeding $100,000. This creates a perverse incentive: flawed research carries high rewards (citations, jobs) with minimal risk of exposure, as verification is prohibitively expensive or impossible without code. While industry researchers at well-resourced labs may rely less on papers due to internal experiments and resources, academic and early-career researchers remain heavily dependent on publications for PhD applications, faculty positions, and entry into top labs. This creates a paradoxical system where papers are increasingly distrusted as reliable knowledge sources yet retain their gatekeeping value in career advancement. The situation underscores a critical need for systemic reforms to ensure scientific integrity and reproducibility in AI research.

marsbit34 хв тому

OpenAI Researcher: We Don't Read Papers Anymore

marsbit34 хв тому

Торгівля

Спот
活动图片