Inspired

I was an engineer. This proved invaluable as I navigated the complex and fast-paced world of financial trading. After chatgpt showed up, I experimented a lot with AI coding. This experience made me quickly improve in quantitative trading on crypto.

However, this is not for the long term because traditional financial institutions are currently limited in movement by law. But, Bitcoin ETFs are going to pass. If this happens, traditional money will flow in here, and the profit will be hard to earn.

That is, I need to change to another field at that time. This inspires me that I need to keep changing to different fields and do Dimensional Reduction Attack.

Doing “Dimensional Reduction Attack” Across Fields

For example, I was an engineer and transferred to use AI coding; the same, I use AI for drawing. Traditional drawing needs lots of references, and finding reference images requires fatigue search images. (Pinterest is a good website with an algo that can find similar images)

And converting to stable diffusion is faster to generate ref images. When I was taking Krenz courses, my homework, when drawing, my references were generated by AI, which is very high efficient. And after GPT-4 with DALL·E 3 can draw, the prompt is more convenient to generate images by DALL·E 3.

By this idea, I should fast move to modeling after finishing Krenz courses (4-dan = rank C to D = 60%) that the 4-dan drawing ability will provide deep understanding to make modeling as Dimensional Reduction Attack. I need to keep in mind that not to involution.

And in the future, after modeling with 4-dan ability, then I need to fast move on to concept art. That is, using 60% modeling ability and 60% drawing ability to achieve concept art.

And next, when I achieve 60% concept art ability, I need to remind myself not to go to involution, then fast change to learn movie storyboard. And after 4-dan storyboard, immediately convert to narrative storytelling.

The Evolution of Skills and Competitive Strategies

The idea of a “dimensional reduction attack” in new fields is fascinating. As complexity in a domain increases, the barrier to entry rises, leading to fewer people being able to make the transition effectively. This scenario has been a consistent aspect of my career transitions. Each new field I’ve entered has presented its unique challenges, requiring a depth of understanding that only comes from thorough learning and application.

This is like UFC and One champion: Demetrious Johnson, he has every fighting skill such as wrestling, boxing, BJJ, Muay Thai, etc., that he can force his enemy to battle in the field that his enemy is unfamiliar with. That is, we need to hold many weapons to force fights on others’ cannot.

Change the Theory Concept of AI Make People More Closer There is a theory that AI makes everything up to 6080% that people need to compete at 9095% to win others.

But now I have another theory, AI makes the upper bound higher.

Before the AlphaGo era, humans pushed Go theory by linear speed to such 3400 that the best Go player’s ELO is 3400. However, AlphaGo-Lee version is 3800, and Master is 4800, breaking the linear speed of improvement of Go theory. AI enormously pushed the upper bound to 4800, and even AlphaGo Zero is 5100. After humans saw AI playing Go, they also learned the method and concepts. Some people say that the distance between pro and amateur is closer and pros are closer, such as the top 500 players are higher than ELO 3000.

But in another view, the 1st Go player, Shin Jinseo, is ELO 3868, and the 2nd Go player is ELO 3698. The distance is even more than 100. Incredibly large.

That we can say AI makes the upper bound higher.AI makes us push to an incredible high level.