I never realized this, but move fast and break things is literally only good for companies who are growing. Move fast is great when you need to catch up. Break things is how you get the cybertruck. I’m not optimistic with any org taking that ethos knowing when to selectively apply it.
Another way to think about is simply risk/reward ratio. When you are behind, or small, or just starting out, "breaking things" has almost infinite upside, little downside. The big question, in the context of AI adoption, is which industry, or which use case within an industry has that r/r profile.
"the neoclouds like CoreWeave or Crusoe, would only build out large clusters if they have at least a three-year commitment (...) take out a mountain of debt from the capital market to fund the purchase of GPUs and energy (...)"
That's why I was intrigued by Voltage Park joining the NAIRR program. Their capital structure is different from other debt-ridden neoclouds, though.
In healthcare the very significant and annual-training-re-emphasized consequences of corporate officer personal liability for data exposures under HIPAA, especially if such was traced back to “we decided to try out this new AI thingee”, will result in “‘No’ is the safest course”.
HIPAA and similar would need carve outs to get healthcare on this train.
I never realized this, but move fast and break things is literally only good for companies who are growing. Move fast is great when you need to catch up. Break things is how you get the cybertruck. I’m not optimistic with any org taking that ethos knowing when to selectively apply it.
Another way to think about is simply risk/reward ratio. When you are behind, or small, or just starting out, "breaking things" has almost infinite upside, little downside. The big question, in the context of AI adoption, is which industry, or which use case within an industry has that r/r profile.
"the neoclouds like CoreWeave or Crusoe, would only build out large clusters if they have at least a three-year commitment (...) take out a mountain of debt from the capital market to fund the purchase of GPUs and energy (...)"
That's why I was intrigued by Voltage Park joining the NAIRR program. Their capital structure is different from other debt-ridden neoclouds, though.
“According to Artificial Analysis, the top four open source AI models are all from Chinese labs. “
I’m scared to look at the top 10.
In healthcare the very significant and annual-training-re-emphasized consequences of corporate officer personal liability for data exposures under HIPAA, especially if such was traced back to “we decided to try out this new AI thingee”, will result in “‘No’ is the safest course”.
HIPAA and similar would need carve outs to get healthcare on this train.