It is relatively easy to teach even the dumb AIs now, how to solve any current problem, and any job. ANY

Simple Pendulum Hamiltonian Equation | Hamilton Equation | bsc 5th semester physics
Shane Sir at https://www.youtube.com/watch?v=l570EfViEgY

There are about 5.4 Billion humans using the Internet now, and many of them can understand English and mathematics and physics.

Do you know your viewers? Do you know how many places Hamiltonians, Lagrangians and principle of least action are used in the world? The new AIs can memorize and solve the equations you are showing, much better. I suggest you learn and show your viewers how to use AIs to solve these problem – exactly. How to verify the results. How to translate and share them in all human languages.

There are about 8.2 Billion humans now and roughly two Billions are from 4-24 years old, and learning for the first time. If you teach them only to memorize and write on paper, they are not going to be able to compete in a world where most professionals will use fairly intelligent computer tools for most everything. If they learn science technology engineering mathematics computing finance government, organizational skills, topic skills (STEMC-FGOT) — and do not know how to get the computer to do it for them, they will not get the best jobs, not be able to compete as a nation, or school, or group.

Show real pendulums, show masses and spring, show waves in water, show vibrations and bending, show things going down a hill, or up a hill. Show things in the real world, and encourage people to transcribe that into sentences an AI can covert to valid mathematical and computer models.

Since most every STEMC-FGOT skill has been studied and improved hundreds of millions of times and much is already in the form of software, apps, APIs, models, online tools – with countless sensors to calibrate and improve the global resources better than a human can. Fairly quickly the AIs will be trained to learn (memorize better than a human) how to run computers better than humans.
 
A human now has a slight advantage over computers because the most marketed AIs are not allowed to program and run computers, or to work unsupervised long enough, and with enough memory, to solve hard problems that humans are doing. That is more by omission than those companies being nice to humans.

You are telling your viewers to “practice it 5 times, or many times”. But you need to give them tools and a framework so they do it. Not just a few times, but thousands of times without error. I think humans are going to be devalued more than they are now. But ones who work on human problems, who understand the world, who can work on real things are likely going to do better than ones who are taught only to memorize — never being shown the real world.

Filed as (It is relatively easy to teach even the dumb AIs now, how to solve any current problem. ANY)

Richard Collins, The Internet Foundation

Richard K Collins

About: Richard K Collins

The Internet Foundation Internet policies, global issues, global open lossless data, global open collaboration


Leave a Reply

Your email address will not be published. Required fields are marked *