This looks like a systemic issue and not just Backblaze : because we’ve layered cloud on top of cloud and call it backup but it’s very linked. When a provider changes the rules then a big part of that safety just disapears and its the same pattern in finance: looks diversified but it isn’t

Interesting that global imbalances have never really disappeared but they were just hidden by low rates : cheap liquidity made deficits easy to finance and reduced the pressure to adjust ; now with higher rates those imbalances start to be important again bacause debt becomes more expensive, the capital flows become more selective and currency pressure is increasing

This looks to be more than just a security bug and rather an incentive problem because you can buy trust with plugin installs numbers and reputation but there’s no mechanism to reprice that trust after the ownership gets changed so the attackers just buy the distribution and monetize it later and that makes this kind of attack economically rational, so it gets reproduced often

Cases are very interesting but most rely a lot on hydro and specific geography : the real shift is not exactly 100% renewables in small systems but maybe more whether large industrial economies can replicate this with maintaining grid stability and lower costs

Good writeup seems like it’s not really the big model against the small one anymore and if smaller models can do most of the job once the context is smaller then it’s more about the system around them and the expertise ...

Interesting how this history is about the edge cases and the unlikely risks that turn into real incidents. the systems scale faster than what we think about their safety.

Some shift looks already happening : scaling AI used to be about better models and more compute but now it’s getting about power cooling and physical limits.

Maybe the bottleneck is shifting from compute to energy and capital ; at some point it stops being a software problem and starts looking like infrastructure power land cooling... Just feels like the constraint is moving down the stack

these transitions looks very uneaven : at macro level it looks like progress but at individual level it can take years of instability before things settle again ; that gap between the aggregate story and experience is probably bigger than we think

Compute stopped behaving like a consumer good and started behaving like an infrastructure; the prices went from competitive cycles to higher while the performance kept compounding and that’s usually what happens when something becomes a bottleneck for entire industries and not just for end users so the gap between what people use and what’s at the frontier says it all.