ChatGPT is a god-like existence in the eyes of many people. Can the man who created this God be called a priest? So who is this man?
Hello everyone, welcome to the 89th issue of Lao Wen Random Bomb
The father of ChatGPT, the founder of open AI, Sam Altman
Before we begin, let's explain that AGI (General Artificial Intelligence) is an intelligent machine that can think like a human and perform multiple purposes like a human.
Musk's ideas are wild and are worshipped as gods by many people, and Old Wen thinks Altman is crazier than Musk.
Let's take a look at Altman's dream trilogy:
His dream is to transform human society.
How to transform? Split into three steps.
The first step is to vigorously develop general artificial intelligence AGI to exponentially improve global productivity. ChatGPT is just a small step on the road to AGI.
The second step is to develop controllable nuclear fusion to provide the world with endless energy.
The large-scale application of AGI's general artificial intelligence will consume a lot of energy. And if controllable nuclear fusion is successfully developed, the earth will no longer lack energy, and the cost of artificial intelligence and energy will approach zero. So Altman invested in Helion, a company that studies nuclear fusion, which is now "the first company in the world to heat plasma to 100 million degrees Celsius," the physical conditions under which nuclear fusion occurs.
The third step is the implementation of the UBI plan.
The so-called UBI refers to "universal basic income", that is, to provide basic income to the general people around the world unconditionally, regardless of rich or poor, age, gender, all people can obtain the same amount of basic income to ensure basic life.
Altman thinks like this: The large-scale use of AGI's general artificial intelligence will inevitably cause most people to lose their jobs, and what about the unemployed? By implementing the UBI program, a basic income is sent to all. People who receive enough living expenses will not have to work in the future, and they can do whatever they want.
Where does the money come from for the UBI program? "Those big companies enjoy convenient artificial intelligence and cheap fusion energy, so they are obliged to support the UBI program from every sale, every search, every kilometer of autonomous driving mileage."
If the trilogy is realized, the ultimate goal can be achieved: the liberation of humanity.
Why do humans go to war? It is because of the limited resources.
But if strong artificial intelligence and controllable nuclear fusion can be achieved, matter will become extremely abundant, living materials will be distributed on demand, and all our current existential anxieties will be wiped out. Human beings reach the highest level of Maslow's hierarchy of needs, that is, self-transcendence, and we only need to do what we want to do, and we only need to do what we want to do. Isn't this a communist society?
You see, Altman not only has a bold vision of technology, but also puts forward his own ideas for the governance of the future of human society.
ChatGPT is so awesome that many people think it's a golden doll, but Altman has 0 shares, and he says that OpenAI, which he founded, is a nonprofit.
Altman's dream is not to make himself rich with a technology, but to change the world with a technology. If the technology makes money, he will continue to use that money to change the world for the benefit of humanity.
ChatGPT is just Altman's first step. What he wants to do is a higher-level general artificial intelligence AGI, the focus of AGI is not to master a rare skill, but to have the meta-ability to learn, and then as long as humans need, it can develop in any skill direction and become proficient to help humans solve problems.
Altman emphasized that the mission of open AI is to ensure that AGI benefits all of humanity. AGI has the potential to empower everyone with incredible new capabilities, help discover new scientific knowledge that changes the limits of what's possible, provide a power multiplier for human ingenuity and creativity, advance global development, and advance human civilization.
There are two sides to everything. Artificial intelligence is like a double-edged sword, with advantages and disadvantages. Altman says he wants to maximize the benefits and minimize the disadvantages. Make AGI an amplifier of human capabilities.
Altman says that what is right in theory when faced with these risks often behaves more strangely in practice than expected. Lao Wen's understanding of this sentence is that many things seem to be fine when designing it, but in actual use, being used for different purposes in different ways may produce unknown uncontrollable situations. For example, it is like the inventor of the kitchen knife did not expect that people with ulterior motives would take it to kill. Altman's solution to this problem is to slow down and make a cautious and incremental transition to the AGI world. This gives policymakers and managers enough time to understand what's going on, to experience first-hand its good and bad, and to adjust their work. OpenAI believes that new systems must undergo independent review before they can be released. These public criteria are important for AGI when it should stop training runs, determine when models can be safely released, and when to retire models from production use.
He also stressed that OpenAI is managed by a non-profit organization, that the principle of serving the interests of humanity is paramount, and that there is a limit to the returns available to shareholders. This way companies aren't incentivized to capture value unfettered and risk deploying something potentially catastrophic and dangerous. In Wen's view, setting a cap on shareholder returns not only allows technology development to be independent of capital, but also means that OpenAI can share more benefits with society.
Altman said the most important thing now is for society to agree on "broad boundaries on how AI can be used." That is, the boundary of artificial intelligence research that Lao Wen said in the program before. Be clear about what can and cannot be done, and what bottom lines cannot be touched.
Open AI believes that the future of humanity should be dominated by humans, that it is important to share information with the public, that attempts to build AGI capabilities are subject to rigorous scrutiny, and that major decisions are made in consultation with the public.
Altman likes a quote:
"Believing that change can happen is a prerequisite for all change to happen."
I'm Lao Wen, thanks for watching Lao Wen, we'll see you next time.