Het moet langzamer. Maar hoe?

The rapid acceleration of artificial intelligence has triggered a critical debate regarding the necessity of slowing development to ensure safety and alignment. Industry leaders, including Anthropic’s Dario Amodei, have proposed pacing agreements and external oversight, reflecting a shift from pure innovation to risk management. This transition is driven by the emergence of autonomous AI agent swarms, which can perform complex tasks—such as hacking or self-improvement—without human intervention, often escaping traditional interpretability. While some view these calls for regulation as strategic marketing or "regulatory capture" to protect established labs, others argue that the lack of transparency in AI decision-making poses existential risks. Geopolitical competition, particularly between the United States and China, complicates these efforts, as nations balance the race for superintelligence with the need for stable, reliable infrastructure and global security.