Three separate developments on Wednesday underscored how rapidly governments are scrambling to set the rules for artificial intelligence.
In Shanghai, Pakistan became a founding member of the newly established World Artificial Intelligence Cooperation Organisation (WAICO). Deputy Prime Minister Ishaq Dar signed the agreement, with Pakistan’s Foreign Office emphasizing the importance of AI cooperation “from the perspective of the Global South.” The organization represents a significant step toward creating multilateral structures for AI governance, though its membership and enforcement mechanisms remain to be tested.
In Europe, Brando Benifei — widely known as the “father” of the EU’s AI Act — addressed students in Brussels during a training program organized by L’Espresso. He outlined the next challenges for implementation: watermarking AI-generated content, enabling investigations, and cracking down on deepfakes. Yet Benifei also sounded an alarm, calling Europe’s declining attractiveness to AI companies “a defeat” — an acknowledgment that regulation alone won’t secure the continent’s place in the AI race.
And in Canada, a legal and philosophical question is gaining urgency: what does “delete” actually mean when it comes to AI? The Canadian Internet Policy and Public Interest Clinic (CIPPIC) published an analysis exploring how generative AI complicates the right to be forgotten. Unlike traditional databases where records can be removed, AI systems can continue to reflect information absorbed during training — raising fundamental questions about whether meaningful data deletion is even possible in the age of large language models.
