The UAE President attended a high-level session focused on artificial intelligence, joined by various ministers and officials. This gathering underscores the UAE's strategic focus on AI development and integration across national sectors. The discussions likely addressed national AI initiatives, policy frameworks, and future implementation plans. Why it matters: The direct involvement of the head of state and key ministers signals a strong national commitment to advancing AI capabilities and leadership within the region.
Researchers at MBZUAI have introduced Video-R2, a reinforcement learning approach to improve the consistency and visual grounding of reasoning in multimodal language models. Video-R2 combines timestamp-aware supervised fine-tuning with Group Relative Policy Optimization (GRPO) guided by a Temporal Alignment Reward (TAR). The model demonstrates higher Think Answer Consistency (TAC), Video Attention Score (VAS), and accuracy across multiple benchmarks, showing improved temporal alignment and reasoning coherence for video understanding.
Researchers at MBZUAI introduce "Interactive Video Reasoning," a new paradigm enabling models to actively "think with videos" by performing iterative visual actions to gather and refine evidence. They developed Video CoM, which reasons through a Chain of Manipulations (CoM), and constructed Video CoM Instruct, an 18K instruction tuning dataset for multi-step manipulation reasoning. The model is further optimized via reinforcement learning with reasoning aware Group Relative Policy Optimization (GRPO), achieving strong results across nine video reasoning benchmarks.
UAE-based sovereign AI chip company Mastiska has raised a $10 million seed round. The funding will be used to develop sovereign AI chips tailored to the specific needs of the UAE and wider region. Mastiska aims to address concerns around data privacy and security by providing locally-controlled AI infrastructure. Why it matters: This investment signals the UAE's commitment to building its own AI capabilities and reducing reliance on foreign technology.
The paper introduces FanarGuard, a bilingual moderation filter for Arabic and English language models that considers both safety and cultural alignment. A dataset of 468K prompt-response pairs was created and scored by LLM judges on harmlessness and cultural awareness to train the filter. The first benchmark targeting Arabic cultural contexts was developed to evaluate cultural alignment. Why it matters: FanarGuard advances context-sensitive AI safeguards by integrating cultural awareness into content moderation, addressing a critical gap in current alignment techniques.