The UAE's extensive utilization of American Graphics Processing Units (GPUs) for its artificial intelligence development has established US technology as a "dominant standard" in the region. This observation was made by Michael Kratsios, former US Chief Technology Officer and AI adviser to Donald Trump. The reliance highlights the critical role of hardware supply chains in shaping global AI capabilities. Why it matters: This underscores the geopolitical implications of technological dependency and the strategic advantage held by nations controlling essential AI infrastructure.
Hamad Bin Khalifa University (HBKU) has released Fanar 2.0, the second generation of Qatar's Arabic-centric Generative AI platform, built entirely at QCRI. The core of Fanar 2.0 is Fanar-27B, which was continually pre-trained from a Gemma-3-27B backbone using 120 billion high-quality tokens and only 256 NVIDIA H100 GPUs. Fanar 2.0 includes capabilities like FanarGuard, Aura, Oryx, Fanar-Sadiq, Fanar-Diwan, and FanarShaheen for moderation, speech recognition, vision understanding, Islamic content, poetry generation, and translation. Why it matters: This shows that sovereign, resource-constrained AI development in the Arabic language is possible, producing competitive systems in the region.
Oman Data Park (ODP) has deployed groundbreaking NVIDIA H200 GPUs to revolutionize its AI infrastructure. This upgrade is set to significantly enhance the processing capabilities available for artificial intelligence workloads within Oman. The move positions ODP to offer more advanced and powerful computing resources to its clients and partners. Why it matters: This deployment could substantially boost Oman's AI ecosystem, attracting advanced AI projects and fostering innovation in the region.
KAUST researchers have been selected as finalists for two ACM Gordon Bell Prizes for high-performance computing. One project used NVIDIA GPUs to enhance genetic studies from the UK Biobank, achieving 133x speedup over existing software. The other developed an exascale climate emulator with higher spatial-temporal resolution than current models, demonstrated on supercomputers like Shaheen III. Why it matters: The recognition highlights KAUST's strength in high-performance computing research and its contributions to both genetic analysis and climate modeling.
The paper introduces Sparse-Quantized Representation (SpQR), a new compression format and quantization technique for large language models (LLMs). SpQR identifies outlier weights and stores them in higher precision while compressing the remaining weights to 3-4 bits. The method achieves less than 1% accuracy loss in perplexity for LLaMA and Falcon LLMs and enables a 33B parameter LLM to run on a single 24GB consumer GPU. Why it matters: This enables near-lossless compression of LLMs, making powerful models accessible on resource-constrained devices and accelerating inference without significant accuracy degradation.
KAUST researchers collaborated with the Paris Observatory and the National Astronomical Observatory of Japan (NAOJ) to develop advanced Extreme-AO algorithms for habitable exoplanet imaging. The new algorithms, powered by KAUST's linear algebra code running on NVIDIA GPUs, optimize and anticipate atmospheric disturbances. The implemented Singular Value Decomposition (SVD) algorithm won an award at the PASC Conference 2018 and is used at the Subaru Telescope in Hawaii. Why it matters: This advancement enhances the ability to image exoplanets, potentially leading to breakthroughs in the search for habitable planets using ground-based telescopes.
Core42, a G42 company specializing in sovereign cloud and AI infrastructure, announced a strategic partnership with Solutions+, an AI-driven shared services company for the MIC Group, at Make it in the Emirates in Abu Dhabi. Core42 will serve as the foundational infrastructure provider, offering GPU compute and large language model access via its Compass API, while Solutions+ will be the primary implementation and data services partner, leveraging this infrastructure for its WEAVE AI platform. The agreement establishes a framework for delivering secure and scalable AI infrastructure across MIC Group companies and Abu Dhabi government entities, with broader scope for government and private sector clients. Why it matters: This partnership significantly advances the UAE's sovereign AI capabilities and accelerates enterprise AI adoption by combining robust infrastructure with specialized implementation expertise for critical national sectors.
MBZUAI researchers Nils Lukas and Toluwani Samuel Aremu will present a paper at ICML 2025 demonstrating the vulnerability of current watermarking techniques in LLMs. Their research shows that adaptive paraphrasers can evade detection from watermarks with negligible impact on text quality, costing less than $10 of GPU compute. The attack involves fine-tuning a small open-weight model to rewrite sentences until surrogate keys no longer trigger detection. Why it matters: This work highlights critical weaknesses in current AI provenance methods, suggesting the need for more robust watermarking techniques to maintain trust in the authenticity of AI-generated content.