But if you listen to Levels' interviews, especially before his Twitter stardom, you will see that he always promotes the find your audience and build to them approach.
It depends, like here in Finland, there is lots of farmland and active farms, but most is at https://en.wikipedia.org/wiki/Humid_continental_climate , Norway also has a lot of farms in subarctic areas like this in the article, but Norway is substituting farming very heavily even European standards.
The term comes from John Doerr https://www.youtube.com/watch?v=n6iwEYmbCwk . But Altman kicked most missionaries out during corporate turmoil in 2023, so not sure where this comes from.
You probably are not a customer as a decision maker in a big traditional company/organization. MS is obfuscating on purpose so they can say in sales decks that if you buy this, you get all these copilots and your Fortune 1000 business is AI-proof. What they are left out is that not every copilot is equal.
The issues with Texas and the UK are that their grids are relatively isolated. Like here in Finland, we have a bigger share of renewables than the UK or Texas, but electricity is still cheaper than UK and pretty much the same as in Texas.
Pretty much all the best stuff is self-published first weekly basis on sites like Royal Road, and then some mainstream publisher will pick it up when it gains momentum.
I think we already hit somekind of performance wall begin of this year. It feels that models are now balancing between rule following and agentic case and general stuff. eg Claude 4 sonet just feels better in Cursor and follows rules very well, and same time it gets equal or worse scores in benchmark against 3.7 Sonet.
You still need "human in the loop" because with simple tasks or some tasks that have lots of training material, models can one-shot answer and are like super good. But if the domain grows too complex, there are some not-so-obvious dependencies, or stuff that is in bleeding edge. Models fail pretty badly. So you need someone to split those complex tasks to more simpler familiar steps.
I think it is more like an NBA situation where a single superstar player can pull up the whole team just look Nuggets but still most likely thing is that the Cavs, Boston, or OCK wins the championship because those teams have a superstar player + a great team.
I think it goes the same way as compilers so bits -> assembly -> c -> jvm and now you don't need mostly care what happens at lower levels because stuff works. With AI we are now basically the bits -> assembly phase so you need to care a lot about what is happening at a lower level.
Was it already happening when platforms started supporting stuff like Iceberg? But is kinda nice to see things like Snowflake have definitely their place on the ecosystem but too often at margins especially with huge workloads Snowflake creates more issues than solves them
Kinda interesting to see where the moat is in AI space. Good base models can always distilled when you have access to API. System prompts can get leaked, and UI tricks can be copied. In the end, the moat might be in the hardware and vertical integration.