AI 模型和平台
在迪拜AI节上正式缩放AI 发布 UAE实体
在 Google 上的首选来源中添加 Unite.AI
缩放AI于2026年10月26日正式发布其阿联酋实体，在迪拜AI节上宣布搬迁，并将新业务设在迪拜国际金融中心(DIFC).
在Scale的公告中，称发布公司连同它于2026年10月早些时候与应用的AI公司Presight签署的谅解备忘录，标志着它在该国工作的一个新阶段。
称规模与世界领先的前沿AI实验室合作，并开发评估与基准测试，帮助业界衡量先进模型的表现，设立DIFC实体的称公司给予其长期存在的基础，并支持其承诺在未来几年与阿联酋机构、伙伴和人才一道工作。
评价、治理和技能
该公司在实务中介绍了这种做法的三个要素。 第一个是评价：确定一个系统必须可靠地做什么，然后在发布之前和之后根据该标准进行测试。 第二是治理：从一开始就将监督建设成一个系统，包括在决策需要时进行人文审查。 第三个是技能：培训当地团队操作、评价和改进系统本身，因此能力留在该国。
比例表指出了该区域其他地方的记录，称它在卡塔尔培训了近6 000人或提高了他们的技能，而且卡塔尔在卡塔尔建造的应用程序被公众日常使用。 该公司还将主权视为对数据、部署、模型和治理的控制和选择，包括随着需要的发展而调整或改变系统的自由。
展望备忘录
2026年10月8日，在阿布扎比的AI Everything活动期间，由Presight首席操作官Adel Al Sharji博士和EMEA负责人Jules de Boinville在"AI"上签字。 在Presight宣布协议时
该协议提出了计划，探索建立代理AI应用的战略合作，加强部署在复杂，任务关键环境下的AI的性能，可靠性和治理，并加快在整个区域负责任地采用AI. 在它之下，称公司将共同探索在阿联酋和其他共同商定的市场上为政府和企业客户设计、建造和完善AI解决方案，成功的产出有可能成为阿联酋和整个区域的政府实体的可重复的报价。
称公司打算将agent开发与评价框架相配对，评价框架一旦在现场运营环境中运行，就不断评估AI agents的准确性，可靠性和性能。 他们还将探索水、电力、保健和运输等关键部门的端到端剂复原力解决方案，并审查模拟如何模拟复杂的业务和紧急情况，在实际世界部署之前测试决策过程，并在各个关键系统中加强备灾能力。 根据公告，评价指导每个agent的设计和部署，确定它必须完成什么，哪些错误是不可接受的，以及必须保留哪些人的权威，然后在系统运行时核实性能是否维持。
Al Sharji 称 AI的真正价值将用它所促成的结果来衡量，这些结果将嵌入支撑现代国家的现实世界体系中，并描述普雷望是阿联酋的主权情报设计师专注于将情报嵌入关键基础设施，从而强化人类决策，维护业务控制，并大规模交付可衡量的成果。
DIFC的AI校园和国家AI框架
“Diubai AI Campus at DIFC”在其公告中引用，由迪拜王储Sheikh Hamdan bin Mohammed bin Rashid Al Maktoum于2024年5月18日宣布成立，作为《迪拜人工智能通用蓝图》第一阶段的一部分，这是一个发布方案，旨在加速在所有相关部门采用AI。 在DFC的就职典礼上发布中，中心称第一阶段完成，超过75个企业在一万平方英尺的校园注册，第二阶段将延伸出10万多平方英尺，并预定吸引500多家公司，到2028年创造出3000多份工作。
DIFC将校区描述为2023年首个宣布，是MENA地区最大的专用集團AI和先进技术。 校区称中心提供基于DIFC独立法律框架的专用AI许可，并便利与包括Amazon网络服务，HP,Microsoft,Oracle和Nvidia在内的公司建立技术伙伴关系。
在国家一级，阿联酋内阁于2019年4月21日通过了"2031年国家人工智能战略"，旨在到2031年将该国定位为AI的全球领袖。 该战略的八项目标包括重申阿联酋作为全球AI中心的地位，为未来的工作吸引和培训人才，提供数据驱动的基础设施以支持AI实验并优化AI治理和监管。 执行工作由酋长国人工情报和数字交易理事会与地方和联邦实体协调监督，执行工作的第一阶段优先考虑资源和能源、后勤和运输、旅游和招待、保健和网络安全。
《阿联酋人工智能开发和使用宪章》规定了目标，包括确保以道德和负责任的方式使用人工智能，保护隐私和数据安全，并加强其使用中的透明度和问责制。 它所宣布的原则包括安全、算法偏差、数据隐私、透明度、人的监督以及治理和问责制，宪章将人对AI的判断和人的监督描述为不可替代的。
本文由 AI宝库自动抓取并翻译，版权归原作者所有；如需查看原文请点下方链接。
AI Models & Platforms
Scale AI Formally Launches UAE Entity at Dubai AI Festival
Add Unite.AI to your preferred sources on Google
Scale AI formally launched its UAE entity on October 26, 2026, announcing the move at the Dubai AI Festival and basing the new operation at the Dubai International Financial Centre (DIFC).
In Scale’s announcement, the company said the launch, together with a memorandum of understanding it signed with applied AI company Presight earlier in October 2026, marks a new phase of its work in the country.
Scale said it works with the world’s leading frontier AI labs and develops evaluations and benchmarks that help the industry measure how advanced models perform, and that it will bring the same rigor to its work with UAE institutions: systems that operate reliably in Arabic, meet local governance requirements and keep people in control. The company said establishing the DIFC entity gives its presence a permanent footing and supports its commitment to work alongside UAE institutions, partners and talent for years to come.
Evaluation, Governance and Skills
The company described three elements of that approach in practice. The first is evaluation: determining what a system must reliably do, then testing it against that standard before launch and after. The second is governance: building oversight into a system from the start, including human review where decisions require it. The third is skills: training local teams to operate, evaluate and improve the systems themselves, so the capability stays in the country.
Scale pointed to its record elsewhere in the region, saying it has trained or upskilled nearly 6,000 people in Qatar and that applications it built there are in everyday use by the public. The company also said it views sovereignty as control and choice over data, deployment, models and governance, including the freedom to adapt or change systems as needs evolve.
The Presight Memorandum
The memorandum of understanding with Presight was signed on October 8, 2026, during the AI Everything event in Abu Dhabi, by Dr. Adel Al Sharji, chief operating officer of Presight, and Jules de Boinville, head of EMEA at Scale AI. “The next phase of AI adoption will be defined by how reliably AI systems perform when deployed in critical real-world environments,” de Boinville said in Presight’s announcement of the agreement.
The agreement set out plans to explore a strategic collaboration to build agentic AI applications, strengthen the performance, reliability and governance of AI deployed in complex, mission-critical environments, and accelerate responsible AI adoption across the region. Under it, the companies said they will explore jointly designing, building and refining AI solutions for government and enterprise customers in the UAE and other mutually agreed markets, with successful outputs potentially becoming repeatable offerings for government entities in the UAE and across the region.
The companies said they intend to pair agent development with evaluation frameworks that continuously assess the accuracy, reliability and performance of AI agents once they are running in live operational environments. They also said they will explore end-to-end agentic resilience solutions across critical sectors such as water, power, healthcare and transportation, and will examine how simulation can model complex operational and emergency scenarios, test decision-making processes before real-world deployment and strengthen preparedness across critical systems. According to the announcement, evaluation guides how each agent is designed and deployed, defining what it must accomplish, which errors are unacceptable and where human authority must be retained, then verifying that performance holds once the system is running.
Al Sharji said AI’s true value will be measured by the outcomes it enables when embedded into the real-world systems that underpin modern nations, describing Presight as the UAE’s sovereign intelligence architect focused on embedding intelligence into critical infrastructure in ways that strengthen human decision-making, preserve operational control and deliver measurable outcomes at scale.
DIFC’s AI Campus and the National AI Framework
The Dubai AI Campus at DIFC, which Scale cited in its announcement, was inaugurated on May 18, 2024, by Sheikh Hamdan bin Mohammed bin Rashid Al Maktoum, Crown Prince of Dubai, as part of phase one of the Dubai Universal Blueprint for Artificial Intelligence, a program launched to accelerate AI adoption across all relevant sectors. In DIFC’s release on the inauguration, the centre said phase one was complete with more than 75 businesses registered on a 10,000-square-foot campus, and that phase two will extend over 100,000 square feet and is set to attract over 500 companies, creating more than 3,000 jobs by 2028.
DIFC describes the campus, first announced in 2023, as the largest dedicated cluster of AI and advanced technology in the MENA region. The centre said the campus offers a dedicated AI licence built on DIFC’s independent legal framework, and that it facilitates tech partnerships with companies including Amazon Web Services, HP, Microsoft, Oracle and Nvidia.
At the national level, the UAE Cabinet adopted the National Artificial Intelligence Strategy 2031 on April 21, 2019, aiming to position the country as a global leader in AI by 2031. The strategy’s eight objectives include reaffirming the UAE’s position as a global AI hub, attracting and training talent for the jobs of the future, providing a data-driven infrastructure to support AI experiments and optimizing AI governance and regulations. Implementation is overseen by the Emirates Council for Artificial Intelligence and Digital Transactions in coordination with local and federal entities, and the first phase of implementation prioritizes resources and energy, logistics and transportation, tourism and hospitality, healthcare and cyber security.
The UAE Charter for the Development and Use of Artificial Intelligence sets objectives that include ensuring AI is used in an ethical and responsible manner, protecting privacy and data security, and enhancing transparency and accountability in its use. Its stated principles cover safety, algorithmic bias, data privacy, transparency, human oversight, and governance and accountability, and the charter describes human judgment and human oversight over AI as irreplaceable.
Auto-fetched and translated by AI Gem Box. All rights belong to the original author.