4 Comments
User's avatar
James Jamington's avatar

What about now in 2024? Have you changed your views whatsoever?

Expand full comment
Philip Skogsberg's avatar

Good question!

I suppose you're asking given recent advancement in the AI space over the past ~year since I wrote the post(?). Tbh I don't see how any of the fundamental points I made have changed eventhough I'm constantly (positively) surprised by the performance improvement of many AI models, and am eagerly awaiting the launch of GPT 5 for example.

I also suspect we will see some tangible improvements in the usability and performance of full-on autonomus AI agents this and next year. Still, the fundamental issues haven't changed; the lack of a "world-model", hallucinations etc.

What do you think?

Expand full comment
James Jamington's avatar

I don’t know Philip, to me it feels like we are playing with fire. Some people think that when there will be a warning shot then people will start to be more cautious. Apart from the fact that in my opinion we’ve had plenty of warning shots, I don’t know if we will survive a warning shot. Also I think the decision to go on with developments of this kind should be decided in a democratic way as the faith of humanity could be on the line.

Look I also hope that this technology helps people that are struggling, but I wouldn’t be able to live in a world that looses all of human touch and human values. Try to imagine a world where everything can and will be done by machines, even friendships, love, families… I don’t know. Maybe it’s just fever dreams, but for sure I don’t want to be around at the fall of humanity.

Thanks for the answer Phil.

Expand full comment
Philip Skogsberg's avatar

Thanks for the answer!

I'm curious. What are some examples of warning shots you have seen?

Expand full comment