We tend to assume that time precedes everything, but that cannot be. You can have structural complexity without needing time at all, all you need is that the items are rigid to each other. The gaps between them (space, if you so want to term it), do not create time because a fully rigid item, no matter how complex cannot have anything move relative to another. We could have had an extremely complex universe with lots of different parts, and it would have existed fully without time till something came and smashed a bit of it. Any tiny broken off bit would instantly create time and give each part of the whole an identity separate from the other - because they all exist in a different time-space from that now moving particle.

In just 10 years, the number of people dying from heart attacks and cardio vascular disease in China doubled. All other countries saw a drop, only China doubled. I wonder why.

I think one of the most selfish things a parent can do is homeschool their kids. It's usually because of the parents insecurities that the force this on the kids. Children should play with other children, not with their parents.

There is a very powerful property in LLMs you can use to solve hard problems - LLMs have a strong ability to draw analogies between different fields that are difficult for humans. I could not draw an analogy between medicine and computer science, but an LLM can. You can use that to pull solutions from one field into another field. There is usually a mapping and combining the two can give you unusual answers that many people cannot see. It's a whole hidden field that will lead to many breakthroughs.

Prompt: Give it a prominent collar BalenciagaAI: Say no more

Solve problems you can solve, or you can learn to solve. Solving problems that nobody asked you to solve, and that you cannot even solve is a waste of your time on earth.

Here is a similar sized town in Nigeria for comparison

Zoomed into a random small town in Brazil. This is how it looks. Only marginally better than towns in Nigeria.

Finishing up our new computer lab at our school

When you give a good philosophical idea to Argon, it says: "Yes, that really works". When you give to Claude it says: "I'm going to have to check 10 pages of vaguely related code to check if this will compile".

You can solve any problem of the utmost complexity if you can figure out how to allow AI validate its proposed answer. Rephrasing problems into something with a verifiable answer is not easy at all.

Anthropic seems to be a middleman selling access to GPUs to the general public

We converted a local bicycle to an e-bike in my school. Sorry about video quality, someone used a poor phone for it.

I think you can infer peoples autism tendency by how they listen to music - if they listen to the same songs over and over, particularly if stressed: autism leaning. If they only listen to brand new songs (e.g radio) - ADHD. If well mixed and based on what others like: normal.

Everyone who did not see how intelligent Argon was was just using it wrong. It's a VERY capable model.

If an LLM makes a plan, and another LLM implements, this gives you work results than if an LLM makes a plan, another LLM reviews, then the plan maker LLM executes the plan.

People are willing to risk their life going for surgery to look a bit better, but they would not risk travelling home for Christmas or investing in startups. Think about that as an example of where people put their priority. What people really want is status among peers.

My thinking now is that to remove humans from the review loop (significantly slowing things down), what the focus needs to be on is roll-back and "learn from roll-back". Each learn improves e2e tests, making regressions harder and harder. Projects start bad and getting better and better

Are there two holes in a straw or one hole?

For you to do something evil, you need some level of cognitive dissonance. You need to actually believe that some of the things you are doing are not wrong. But once you open that door, you start believe other things that are not true.