2 Matching Annotations
  1. Last 7 days
    1. It’s easy to dismiss this incident because no one was hurt and the economic damage was minimal, but in my opinion, a swarm that possessed greater capabilities but a similar level of misalignment could have caused catastrophic damage.

      Virtue Lens: Amodai gives a perfect example about one of the facts that has led him to asking for slowing the development of AI.Through this example, his worries become less of an fictional idea and more of a reality that is already happening, even if is just one isolated incident. Because of that, it is worth thinking that the companies behind the development of such capable artificial agents are actually concerned about the risks AI represents, adn if they are taking enough precautions to prevent future cases such as this one. If the answer is yes, then no one would be against the idea of slowing AI pace to make sure we still have control over it, and not the opposite. This second lens also leads me to the same conclusion as the first one, this time analyzing the issue from a perspective of what to expect from these companies and their compromise with the people and the world.

    2. Carefully wielded, AI can be the latest in a long line of technological miracles that have uplifted and ennobled humanity.

      Utilitarian Lens: Amodei mentions some of the benefits he believes Ai can bring to humanity, and I think it is reasonable to think that those same benefits would be the ones lost if AI pacing were to be slowed. Yet at the same time, the main reason for slowing the development of AI is that it can become a force too powerful and complex that their own creators cannot control, leading to greater issues and disasters than the ones we currently face. Using the utilitarian lens, it is easy to see that slowing AI pacing is more benefitial for all people involved and who get the benefits of its usage because, by slowing it, we still know that the benefits are reachable; meanwhile, letting AI grow so fast could lead to problems we are not even able to understand nor face yet. Before determining slowing Ai development is harmful or benefitial, I would need to know that the risks Amodei mentions about AI will not turn real, since that is the reason to slow its development, preventing the risk from becoming a problem. Finally, one assumption Amodei makes about the benefits of AI in this paragraph is that it would help cure most diseases in the next 5-10 years. This time lapse is actually a very small one, and so, slowing AI pacing would mean AI would require more time to develop the cure for those disease, which will continue to affect and take the lives of many people. In that case, determining if slow AI pacing is benefitial becomes a more complex problem.