Sam Altman is expecting a child in 2025. Congratulations to Sam!
This market resolves yes if by my assessment Sam Altman becomes more cautious about developing unaligned AGI / ASI after having a child.
After the birth of his child:
If Sam explicitly says that he is more cautious after having a child, this resolves yes.
If there is a clear move to slow things down from OpenAI (eg. Coordinating with the other big labs to agree to a pause, stating that they will not develop any new frontier models until they have done more research on exisiting models) this resolves yes.
If openAI endorses some regulatory proposal that would limit frontier model development this resolves yes.
If otherwise consensus agreement from legitimate sources is that Sam has become more cautious since having a child this resolves yes.
Otherwise if things continue until mid 2026 with none of this or any other move that looks to be an increase in cautiousness from Sam / openAI this resolves no.
1,000Pessoas também estão operando
This is substantially weaker than
"Coordinating with the other big labs to agree to a pause, stating that they will not develop any new frontier models until they have done more research on exisiting models"
He said they're pausing training until they fix the sandbox. That doesn't even involve research on existing models.
Pacing development so society can firm up cyber security defenses is also not concrete, he says they may have to. And it's focused around current models, not AGI/ASI.
This is the formal statement
"We believe that, at some point in the future, AI acceleration for frontier model development may be so high that the world will need to pace the rate of AI advancement.
We hope to contribute to work led by the U.S. government, alongside other labs and the open-source community, to develop the tools and mechanisms that could make that possible."
"We request that the U.S. government support an international effort to develop the technical and governance tools needed to deliberately pace the frontier of automated AI development.”
This is closer but still seems short of actually supporting a particular proposal. At some point in the future we may want to pause, let's see if we can get more details from the government.
@billyhumblebrag Yeah I don't know, if you showed me this blog post a year ago there's no way I wouldn't be pretty shocked and think this lines up with the market.
>We have long believed there should ultimately be an international organization that helps coordinate leading AI efforts to reduce catastrophic risk. Cooperation and shared safety standards are an important part of the path forward, especially because the incentives around commercial and national competition are hard to escape. One goal of such an organization should be to make it possible for the world to take coordinated action, including slowing frontier development when needed, so societal resilience, safety, and alignment can keep pace.
https://openai.com/index/built-to-benefit-everyone-our-plan/ from June 2026
>There are many ways this could be implemented; major governments around the world could set up a project that many current efforts become part of, or we could collectively agree (with the backing power of a new organization like the one suggested below) that the rate of growth in AI capability at the frontier is limited to a certain rate per year.
https://openai.com/index/governance-of-superintelligence/ May 2023
Is the current statement really that much beyond the prior ones?
@AviEisenberg I would say yes because there was a large doubt as to whether or not the previous statements were lip service, whereas this one has stakes, they're actually maintaining that position in a capacity that has an effect on government, they actually decided to restrict internal usage of this model and to stop training using it, etc.
Reminds me of that quote from Tenet (slight spoilers in case you haven't seen it yet):
"My greatest sin was to bring a son into a world I knew was ending. Do you think God will forgive me?"
If Sam thinks there is a good chance this might end civilization as we know it and he knowingly puts a child into the world... oef....
@billyhumblebrag would you say that there ist current any evidence of increased caution? I would say no pretty clearly.
