Certainly! Here’s a rewritten version of your text:

The fear surrounding artificial intelligence (AI) is growing as we approach significant milestones, such as when human-level intelligence might be achieved. Critics argue that this could lead to catastrophic consequences for humanity. For instance, if an AI system surpasses human cognitive abilities, it could pose a direct threat to survival.

One of the most pressing concerns is the immortality of AI models. Once trained, these systems can continue functioning even as their underlying hardware deteriorates or is replaced. This longevity poses a persistent risk, as older models will remain operational long after newer ones are developed and deployed.

To mitigate these risks, Geoff Hinton suggests treating AI like nuclear weapons—something humanity must handle with extreme caution. He emphasizes the importance of international cooperation to ensure that AI development stays under control. Given how interconnected global events have become, even a small oversight could escalate into an unmanageable crisis.

Hinton’s call for action is further underscored by his friend’s advice: "Geoff, you need to speak. They will listen to you. People are blind to this danger." This sentiment highlights the urgency of addressing AI risks before they escalate beyond our control.

In conclusion, the potential dangers and immortality of AI necessitate a proactive approach from all nations. Collaboration is key to ensuring that AI remains a tool for humanity’s benefit rather than a threat.

This version maintains the core ideas while presenting them in a more concise and structured manner.