TL;DR:
- Saudi Arabia and UAE are purchasing Nvidia chips for AI development.
- Gulf nations are striving to lead in AI and boost their economies.
- Concerns arise over the potential misuse of AI tech.
- Saudi Arabia procuring 3,000 Nvidia H100 chips via Kaust.
- UAE was developing its own Falcon model with thousands of Nvidia chips.
- Both nations aim for self-sufficiency in computational power.
- Global tech companies are also vying for scarce Nvidia chips.
- Western AI leaders express ethical and safety concerns.
- Gulf states leveraging petrodollar windfalls for AI advancements.
- Abu Dhabi is recognized for foresight in AI development.
Main AI News:
In a bid to fortify their positions as AI trailblazers, Saudi Arabia and the United Arab Emirates (UAE) are strategically amassing substantial quantities of high-performance Nvidia chips—an essential component in the construction of artificial intelligence software. This push has ignited a global race for AI supremacy, amplifying demand for these coveted Silicon Valley assets. These Gulf powerhouses have publicly declared their ambitions to spearhead advancements in AI, underpinning their bold economic expansion strategies. However, this fervor also sparks apprehensions regarding the potential misuse of such technological prowess by the autocratic rulers reigning over oil-rich states.
According to informed sources, Saudi Arabia has procured over 3,000 units of Nvidia’s H100 chips, each priced at $40,000. These cutting-edge processors, acclaimed by Nvidia’s CEO Jensen Huang as “the world’s inaugural computer [chip] tailored for generative AI,” were acquired via the esteemed research institution King Abdullah University of Science and Technology (Kaust). Similarly, the UAE has seamlessly secured access to an extensive cache of Nvidia chips, propelling its advancement in AI by independently cultivating its proprietary open-source large language model, christened Falcon. This groundbreaking initiative was realized within the confines of the state-owned Technology Innovation Institute, nestled within Abu Dhabi’s Masdar City.
An authoritative source familiar with Abu Dhabi’s strategic vision conveyed, “The UAE has made a decision that it wants to possess and steer its computational prowess and human resources, to nurture its autonomous platforms and lessen reliance on external entities such as China and the United States.” Undoubtedly, the UAE’s substantial financial reserves, coupled with its wealth of energy resources, augment its capacity to execute this formidable undertaking and allure leading global talent to its shores.
The notable acquisitions of Nvidia chips by these Gulf states, facilitated through their state-controlled entities, align with an international trend where leading tech conglomerates relentlessly compete to secure these scarce chips for AI development. Notably, prominent Chinese tech giants like Tencent and Alibaba are fervently pursuing the procurement of Nvidia’s high-performance chips. Meanwhile, the most advanced Large Language Models (LLMs) remain under the domain of US corporations, including Microsoft-backed OpenAI and Google. These entities stand as primary patrons of Nvidia’s H100 and A100 chips. As revealed by insider sources within Nvidia and its manufacturer, the Taiwan Semiconductor Manufacturing Company, approximate 550,000 units of the latest H100 chips are slated for global distribution in 2023, with a predominant share being allocated to US tech entities.
Transitioning to Saudi Arabia’s trajectory, Kaust is anticipated to receive 3,000 specialized chips, with a cumulative value of approximately $120 million, by the culmination of 2023. This formidable assemblage complements the university’s ongoing initiative to constructing a supercomputer named Shaheen III, which is projected to be operational this year. Notably, this computing behemoth will house 700 Grace Hoppers, Nvidia’s preeminent superchips expressly designed for cutting-edge AI applications. Kaust’s strategic intent is to leverage these chips to engender its bespoke Large Language Model, akin to OpenAI’s GPT-4, which empowers the immensely popular chatbot ChatGPT.
The Saudi LLM finds its genesis within the Provable Responsible AI and Data Analytics lab at Kaust. This think-tank predominantly comprises Chinese researchers, many of whom opted for Kaust due to their exclusion from studying and working in the US post-graduation from Chinese universities on the US entity list. Notably, the UAE has notably championed AI, establishing a dedicated AI ministry in 2017—the first of its kind. As an emblem of its commitment to this endeavor, the UAE has introduced a “Generative AI Guide,” reflective of its ambition to reinforce its stature as a technology and AI innovator on the global stage.
Further bolstering this trajectory, the UAE’s Falcon model, a notable achievement, has been made freely accessible online. Impressively, this model was cultivated by harnessing the power of 384 A100 chips over two months. Acknowledged AI luminaries, including leading researchers and venture capitalists such as Marc Andreessen, have lauded the accomplishments of this model, further underscoring the UAE’s ascendance in the AI domain. As part of its concerted efforts, the UAE government has secured additional Nvidia chips, envisaging an expanded array of LLM-associated applications and cloud services.
Nevertheless, apprehensions linger within Western AI thought leadership and human rights circles regarding the ethical compass and safety protocols that may be absent in the AI solutions developed by these two nations. These concerns gain potency given the documented instances of government crackdowns on human rights defenders and journalists within UAE and Saudi Arabia. Iverna McGowan, Director of the Center for Democracy and Technology’s Europe office, conveys a disconcerting notion that when coupled with AI’s propensity for discriminatory consequences and its potential to fuel unlawful surveillance, the current trajectory is indeed unsettling.
The ascending trajectory of Saudi Arabia and the UAE in the AI arena has been significantly propelled by the influx of petrodollars, triggered by a surge in energy prices in recent times. Both nations expertly oversee some of the globe’s most substantial and dynamic sovereign investment funds, endowed with unprecedented fiscal resources. Recent overtures by representatives of these Gulf state-affiliated funds to Western AI startups, aiming to exchange computing resources for invaluable code and LLM expertise, bear testament to their strategic intent in this AI foray.
OpenAI’s CEO, Sam Altman, underscored Abu Dhabi’s prescience in recognizing the pivotal role of AI during a visit to the region. Speaking in a question-and-answer session held within the financial district, Altman envisaged the Gulf region’s pivotal role in shaping the global discourse on this nascent technology and its accompanying regulatory framework. He remarked that, “There has been discussion about AI, here in particular, in Abu Dhabi, before it was cool. Now everyone is on the AI bandwagon, which we are excited about, but we have a special appreciation for the people who were talking about this when everyone thought AI was not going to happen.”
Conclusion:
Saudi Arabia and the UAE’s concerted efforts to acquire Nvidia chips signal a fierce race in the AI landscape. While their ambitions to become AI leaders are commendable, concerns about ethical considerations and safety protocols emerge. This trend highlights the increasing demand for Nvidia chips globally, intensifying the competition among tech giants. The Gulf states’ utilization of their substantial resources underscores the transformative potential of AI in their economic strategies. Abu Dhabi’s early recognition of AI’s significance positions it as a pivotal player in shaping the technology’s future.
