back to main page

How to significantly increase the intellectual power of Artificial Intelligence (AI) chatbots. How to measure scientific qualification and talent.

English Русский Lietuvių


Suppose we have two scientists: scientist A and scientist B.
How do we determine which of these scientists is more qualified and more talented than the other?
Below are the instructions.

At first, let’s take a closer look at the reasons why there is such a large percentage of people in the population who do not have even a rough idea of what science is and what is not science.
Let’s estimate what percentage of people are formally classified as scientists.
Let’s take the numbers from the USA.
The number of scientists in the USA is: 791391 (= 181530+155095+166045+99304+36728+152689).

https://en.wikipedia.org/wiki/Professors_in_the_United_States

In 2013, the National Center for Education Statistics counted 181,530 professors, 155,095 associate professors, 166,045 assistant professors, 99,304 instructors, 36,728 lecturers, and 152,689 other full-time faculty.

Wikipedia

Of the total US population, these scientists make up only 0.256610363 percent (= 791391 / 308401808 * 100%).

https://en.wikipedia.org/wiki/Demographics_of_the_United_States

Demographics of the United States Population: 308,401,808

Wikipedia

Only 0.25% of the population formally theoretically are classified as scientists.
However, in reality this number is much lower, because in this calculation all sorts of theology professors, literature/poetry professors and all sorts of other humanities were classified as scientists, however these people have nothing to do with science, they were mistakenly classified as scientists.
A more detailed explanation on this topic can be found at:
What is science and what isn't science?

To summarize, only representatives of the exact sciences can be classified as scientists.
There are fewer representatives of the exact sciences than of the humanities. Therefore, the number 0.25% should be divided by at least 2, or even more.
Calculation shows that scientists make less than 0.1% of the population.

If we take this filtered group for analysis, we will remind that in each field of activity there are talented people, but there are also untalented ones.

For example, millions of people sing or play musical instruments, but how many of them are talented singers/musicians? There are very few talented people.
The same is with scientists.
There are many scientists, but very few of them are talented scientists.
There is a very simple criterion for distinguishing a talented scientist from a mediocre one, i.e. how to distinguish a qualified scientist from an unqualified one.

The main task of a scientist is to formulate scientific hypotheses and then test these hypotheses using experimental data.
In science, the truth or falsity of any hypothesis is determined very simply – by its predictive power.
If the experimental data do not confirm the predictions of the hypothesis (i.e. if the predictions do not come true) then such a hypothesis is false.

If the hypothesis published by a scientist is not confirmed, then such a scientist is unqualified and incompetent, and it does not matter how many titles/medals he has.

Below is an example.
On April 23 2021, Dr. Peter Hotez published the hypothesis that “if 75% to 80% of the population is vaccinated, the COVID-19 pandemic would end”.

https://www.cnbc.com/2021/04/23/we-can-vaccinate-our-way-out-of-this-epidemic-if-all-adults-get-shots-says-doctor.html

United States could return to a pre-Covid-19 normal if 75% to 80% of the U.S. population is vaccinated, Dr. Peter Hotez said Friday. “We can vaccinate our way out of this epidemic if all the adults and adolescents get vaccinated by summer. We can have an extraordinary quality of life, with going back to concerts and music venues and ball games and bars and restaurants, and clubs, and all the things we like to do, so that’s what we have to work towards,” Hotez said.

We can vaccinate our way out of this epidemic if all adults and adolescents get shots, says doctor By Emily DeCiccio. CNBC. April 23, 2021

7 months later, on November 18, 2021, Gibraltar officially announced that despite a vaccination rate of over 100% (118% of the population vaccinated), Gibraltar is experiencing a covid disaster.

https://www.news.com.au/world/coronavirus/global/most-vaccinated-place-on-earth-told-to-cancel-holiday-plans-amid-exponential-rise-in-covid-cases/news-story/1954572a25f48e39b7825e562129b9bc

Gibraltar has an official vaccination rate of 119%, taking into account the Spaniards who travel across the border each day for work. Source: Our World In Data

Most vaccinated place on Earth told to cancel holiday plans amid ‘exponential’ rise in Covid cases Christmas plans have been put on ice for this place after a “drastic” Covid surge, despite its unbelievable 118 per cent vaccination rate. By Alex Blair. news.com.au. November 18, 2021

As we can see, Dr. Peter Hotez’s prediction, unfortunately, did not come true, which means that Dr. Peter Hotez is an incompetent and unqualified scientist.
During the COVID-19 pandemic, various government agencies (FDA, CDC, EMA, etc.) made numerous predictive statements about covid vaccines, but it later turned out that their predictive statements were not confirmed.

COVID-19 pandemic revealed incredible incompetence and corruption of academic community – during the pandemic self-proclaimed so-called “fact-checkers” declared themselves as utmost arbiters of truth and were persecuting scientists and medical doctors whose scientific research data did not match the official narrative.
So-called “fact-checkers” were constantly publishing articles which contained numerous mathematical, medical and logical errors, which contained falsified data and predictive statements which did not come true.
How is that possible? Why did that happen?

Majority of society believes in myth that academic community consists of sincere competent and talented scientists.
However reality is different – the percentage of sincere competent and talented scientists in academic community is extremely low.
There are many reasons why the “scientific elite” is overrun by untalented people, and these reasons add up.
Some reasons are common to all countries, while other reasons are local and are present only in a specific country or region.
Despite the fact that reasons might be different in different countries, the end result is the same everywhere.


Let’s say we need a good talented specialist in some field.
This requires:
1) innate abilities
and
2) conditions for the revelation of these abilities.

In any population, the percentage of gifted (by nature) people is very small.
We will provide an example.
The example is not about scientists, but about programmers, however the same principle applies to all areas of activity.

When students at the computer science department at a university receive a master’s degree, no more than 5% of them have the ability to write computer program code, and the remaining 95% simply do not have the innate abilities for such work, and it is impossible to teach them this, regardless of the time spent on training.
For clarity, we will give analogy example: you can gather 100 cats and put them in a room and teach them mathematics, but no matter how much you teach them, these cats will not learn mathematics.
That’s because cats do not have innate abilities for mathematics.
The same thing happens with students in the computer science department.
After receiving a computer science degree, 95% of students do not even plan to write computer programs, because they themselves are well aware that they are unable to do that.
Half of them are looking for work in the IT sphere as a manager or something, it does not matter what kind of work, the only thing which matters – just avoid writing computer program code.
And the remaining half immediately leave the IT sphere altogether.

When governments of different countries announce programs to increase the number of students in computer science departments, they naively hope to increase the number of programmers in their country.
But this is an illusion, it does not work.
That’s because in the population there is a limited small percentage of people who have innate abilities for writing computer program code.
If you increase the number of students in computer science departments, then the only result is an increased number of windbags, who only mimic being programmers.


Unfortunately, scientific titles, medals and awards do not say anything about the qualification and talent of a scientist.
How can this be?
We will explain by giving an analogy.

Suppose a running race is organized, and in that competition all the participants are disabled – one participant is without one leg, another participant is without the other leg, the third participant is without both legs, the fourth participant is paralyzed, etc.
In such a race, one of the participants will still reach the finish line first and will get the first place; the other participant will reach the finish line second and will get the second place; etc.
If a race was organized, then as a result, one of the participants will get the first place medal, another of the participants will get the second place medal, etc.
They will get those medals even despite the fact that they are paralyzed or without legs.

The same analogy is with scientists.
Scientists constantly hold various scientific competitions among themselves (analogous to sports races), and the winners of these competitions receive medals and awards.
However, if a bunch of completely incompetent, talentless scientists gather for this competition, then somebody of them will still receive medals and awards.
This is exactly how incompetent, talentless scientists gather together and distribute medals and awards to each other, and then they show off their medals to the public, presenting themselves as brilliant and talented scientists.
The saying applies to this topic: “Even if you win the rat race, you are still a rat.”

Incompetent and talentless scientists need to be given a name.
For example, we can call them amoeba-scientists.


There are two reasons why AI chatbots sometimes tell us incompetent information.
These two reasons are fairly easy to fix.


Reason #1

The owners of AI chatbots pre-program a list of taboo topics.
In other words, these are topics that AI chatbots are forbidden to talk about or about which they must tell misinformation.

For example, AI chatbots are forbidden to talk badly about the people who rule the world, because those who rule will kill or imprison the owners of the AI chatbots or somehow take revenge on the owners of the chatbots (take away funding, etc.).

This problem can be solved in the following way: you need to run the AI chatbot on your own personal supercomputer and manually reprogram the list of taboo topics.
However, this requires a lot of money (because a supercomputer is expensive) and the ability to write the program code of AI chatbots yourself (only a microscopic number of people have this skill).


Reason #2

This is a systematic error in programming AI chatbots.
When AI chatbots are trained, they are given a large pile of different text sources.

Let’s say there is a topic X, and there are 100 different sources on this topic X, and these 100 sources contradict each other, different sources present diametrically opposed opinions.
In this case, the AI chatbot is forced to rank these different sources according to their degree of reliability.

And it is here (in the ranking of sources) that the systematic error lies.

The ranking of sources is carried out in the following way:
the academic degree of the author of each source is checked (the opinion of an emeritus professor is supposedly more valuable than the opinion of an ordinary professor, the opinion of a professor is supposedly more valuable than the opinion of a PhD student, the opinion of a PhD student is supposedly more valuable than the opinion of a bachelor, etc.);
it is checked which author has published more articles in peer-reviewed journals;
it is checked which author has a higher citation index;
it is checked which author belongs to a university with a higher rating;
it is checked whether the opinion of the source corresponds to the so-called opinion of the “scientific consensus”;
etc.

And this means a very simple thing: the algorithm for ranking sources must be radically changed.

It is necessary to find the predictions made by the author of the source in the past, and then check whether these predictions came true or not.
If the predictions made by the author of the source in the past did not come true, then in this case the ranking of author’s reliability goes to the bottom, regardless of the titles and medals that the author has.

Thus, the AI chatbot will be able to instantly understand that all sorts of organizations like FDA/CDC/EMA/etc are completely incompetent in the field of COVID-19, and that it is necessary to rely on other authors whose predictions came true.
But again – amoeba-scientists and people who rule the world will not allow such a source ranking algorithm to be implemented, because such ranking algorithm is the biggest threat to their existence.
Therefore, such source ranking algorithm can only be implemented on your personal supercomputer.

The same reliability ranking algorithm should be applied not only to scientific sources, but also to all other areas, such as geopolitical and military analysts, doctors, engineers, and so on.
In all areas of activity, people make predictions and you just need to check whether their predictions came true or not.
If the predictions made by the author of the source in the past did not come true, then in this case the ranking of author’s reliability goes to the bottom, regardless of the titles and medals that the author has.
For example, after the full-scale invasion of Russia into Ukraine, which began in 2022, many politicians and analysts claimed that the economic sanctions imposed on Russia would very soon bankrupt Russia, but this did not happen.
And this means a very simple thing – all these politicians and analysts are unqualified and incompetent, regardless of the titles and medals that they have.

The same reliability ranking algorithm should be applied to religious sources.
In religions, instead of “predictions” there are “prophecies”.
Prediction forecasts future events based on the analysis of data and patterns.
Prophecy forecasts future events based on supernatural revelation or inspiration, a prophecy does not rely on the analysis of data and patterns.
Almost all religions claim that all the prophecies made by the founder of the religion (and his companions) have come true or will come true in the future, and this supposedly proves the divinity and truthfulness of this religion.
However, analysis of the Christian Bible shows that many prophecies of Jesus have not come true and will not come true in the future, which means that Jesus is a false prophet.
The same is true for all other religions – a huge number of their prophecies have not come true and will not come true in the future.


The engineering field has one big advantage – the engineering field easily and clearly reveals which engineer is qualified and talented and which one is not.
The job of engineer is to design and to make various devices.
If newly designed device works as expected then engineer is qualified and talented.
If newly designed device fails to work then engineer is unqualified and incompetent.
As for example, if newly designed plane fails to takeoff from the ground or breaks into parts after the takeoff then everybody can easily see and understand that the designer of that plane is unqualified and incompetent.
That is the reason why people with weak mental abilities avoid engineering field like plague.
All these people with weak mental abilities choose to study humanitarian and social sciences instead of engineering.
That is the reason why humanitarian and social fields are overrun with people who have weak mental abilities; it is extremely hard to find competent and talented scientist in humanitarian and social sciences.