Microsoft is investing $1.5 billion in G42 to accelerate AI development in the UAE. This investment includes bringing Microsoft’s latest AI technologies and skilling initiatives to the UAE. The partnership also involves a strong focus on AI safety and security. Why it matters: This represents a significant commitment to the UAE's AI ecosystem and positions the country as a major hub for AI innovation in the region.
This study introduces a Probabilistic Graphical Model (PGM) framework utilizing Pearl's do-operator to causally audit LLM safety mechanisms, specifically isolating the effect of injecting cultural demographics into prompts. A large-scale empirical analysis was conducted across seven instruction-tuned models from diverse origins, including the UAE's Falcon3-7B, as well as models from the US, Europe, China, and India, using ToxiGen and BOLD datasets. The findings revealed a disparity between observational and interventional bias, demonstrating that standard fairness metrics can overestimate demographic bias. Western models exhibited higher causal refusal rates for specific demographic groups, while Eastern models showed low overall intervention rates with targeted sensitivities toward regional demographics. Why it matters: This research highlights the geopolitical nuances of LLM safety alignment and the potential for demographic-sensitive over-triggering to restrict benign discourse, which is particularly relevant for diverse regions like the Middle East in developing culturally-aware AI.
Technology Innovation Institute (TII) and Mila, the Quebec AI Institute, announced a strategic partnership to collaborate on AI safety and next-generation machine learning models. TII will establish a research lab at Mila in Montreal, enabling collaboration between UAE-based researchers and Mila's AI specialists. The partnership aims to translate scientific advances into real-world impact and strengthen the global research ecosystem. Why it matters: This collaboration enhances UAE-Canada scientific ties, positioning both communities for breakthroughs in areas like LLMs and AI safety, aligning with the UAE's vision to become a global AI research hub.
Saudi Arabia is participating in the drafting process of the International AI Safety Report, which is slated for release in 2026. This involvement underscores the Kingdom's commitment to global AI governance and the development of responsible AI standards. The initiative aims to establish unified international guidelines for the safe deployment and use of artificial intelligence technologies. Why it matters: Saudi Arabia's active role in shaping this crucial international document positions it as a key contributor to global AI policy and helps advance its strategic vision for a secure and ethical AI future.
UAE President Sheikh Mohamed bin Zayed Al Nahyan met with OpenAI CEO Sam Altman to discuss opportunities for collaboration in artificial intelligence. They explored the potential of AI to support the UAE's growth across various sectors. The meeting also covered the importance of AI safety and regulation. Why it matters: This high-level engagement indicates the UAE's commitment to being a leader in AI adoption and development, while also highlighting the need for responsible AI governance.
Researchers from MBZUAI and other institutions have developed a new framework called STEREO to improve the safety of text-to-image diffusion models. STEREO uses a two-stage approach: STE (Search Thoroughly Enough) based on adversarial training and REO (Robustly Erase Once) for batch concept erasure. This framework aims to enhance safety without significantly impacting the model's performance on normal queries. Why it matters: The framework addresses vulnerabilities in AI image generation, reducing the creation of inappropriate images while preserving performance on harmless queries.
MBZUAI and École Polytechnique are deepening research collaboration through a Collaborative Research Agreement, focusing on large language models, foundation models for reasoning, and AI applications in biology, health, and AI safety. The partnership builds on a previous MoU and Scholars Exchange Program Agreement between the two institutions. MBZUAI's France Lab has developed Atlas-Chat, a family of open-source LLMs for the Moroccan Arabic dialect Darija, with models including Atlas-Chat-2B and Atlas-Chat-9B. Why it matters: This collaboration strengthens the AI ecosystems in both France and the UAE, fostering joint research efforts and supporting the next generation of AI researchers and innovators, with a specific focus on Arabic NLP.
MBZUAI-based startup LibrAI has launched the Libra-Leaderboard, an evaluation framework for LLMs that assesses both capability and safety. The leaderboard evaluates 26 mainstream LLMs using 57 datasets, assigning scores based on bias, misinformation, and oversensitivity. LibrAI also launched the Interactive Safety Arena to engage the public and educate them on AI safety through adversarial prompt testing. Why it matters: The Libra-Leaderboard provides a benchmark for responsible AI development, emphasizing the importance of aligning AI capabilities with safety considerations in the rapidly evolving LLM landscape.