At 5 a.m. today, Sam Altman, co-founder and CEO of OpenAI, published an in-depth article titled &#34;Three Observations&#34; on his personal blog.The article presents three observations about the AI world: the intelligence level of AI models is roughly equal to the logarithm of the resources used to train and run them; the cost of using AI at a specific level decreases by about tenfold every 12 months, and lower prices lead to more usage; the socioeconomic value created by linearly growing intelligence is super-exponential.The article also emphasizes that OpenAI is fully committed to developing AI Agents, which will eventually be widely used as virtual colleagues in various business scenarios. The performance of basic Agents may be somewhat lacking, but they will still permeate various fields and reshape the world economy.<img src="https://imageproxy.pbkrs.com/https://wpimg-wscn.awtmt.com/1835d47d-aa39-4aaf-bbcf-20add5de76b1.png?x-oss-process=image/auto-orient,1/interlace,1/resize,w_1440,h_1440/quality,q_95/format,jpg" alt="" original-src="https://imageproxy.pbkrs.com/https://wpimg-wscn.awtmt.com/1835d47d-aa39-4aaf-bbcf-20add5de76b1.png"/>Here is the original content:Our mission is to ensure that general AI (AGI) benefits all of humanity.Systems pointing towards AGI are beginning to emerge, and we feel it is necessary to understand the moment we are in. AGI is a somewhat vague term, but generally, we define it as systems capable of solving increasingly complex problems at human levels across many domains.<img src="https://imageproxy.pbkrs.com/https://wpimg-wscn.awtmt.com/71810c65-d491-4a68-b3be-94147d989d1d.png?x-oss-process=image/auto-orient,1/interlace,1/resize,w_1440,h_1440/quality,q_95/format,jpg" alt="" original-src="https://imageproxy.pbkrs.com/https://wpimg-wscn.awtmt.com/71810c65-d491-4a68-b3be-94147d989d1d.png"/>Humans are toolmakers, inherently driven by the motivation to explore and create, which makes the world a better place. Each generation builds upon the discoveries of its predecessors, creating more powerful tools—electricity, transistors, computers, the internet, and the upcoming AGI.In a sense, AGI is just another tool in the progressive ladder built collectively by humanity. But in another sense, the emergence of AGI marks a brand new beginning, making it hard not to feel that &#34;this time is really different.&#34; The economic growth before us looks astonishing, and we can even imagine a world that can cure all diseases, allow people more time with their families, and fully unleash creativity.Perhaps within a decade, the capabilities of everyone on Earth will surpass those of today's most influential individuals.<img src="https://imageproxy.pbkrs.com/https://wpimg-wscn.awtmt.com/8616aaa4-87fa-42a7-ac41-eaaa290662d9.png?x-oss-process=image/auto-orient,1/interlace,1/resize,w_1440,h_1440/quality,q_95/format,jpg" alt="" original-src="https://imageproxy.pbkrs.com/https://wpimg-wscn.awtmt.com/8616aaa4-87fa-42a7-ac41-eaaa290662d9.png"/>We continue to witness the rapid progress of AI development. Here are three observations about AI economics:<ol><li>The intelligence level of AI models is roughly equal to the logarithm of the resources used to train and run them. These resources mainly include training computing power, data, and reasoning computing power. It seems that as long as you are willing to invest any amount of money, you can achieve continuous and predictable returns. The scale law predicting this return is accurate across multiple orders of magnitude</li><li>The cost of using specific levels of AI decreases by about tenfold every 12 months, and lower prices will lead to more usage. You can see this from the change in token costs from GPT-4 at the beginning of 2023 to GPT-4o in mid-2024, where the price per token has dropped by about 150 times over this year and a half.</li></ol>Moore's Law doubles performance every 18 months, changing the world; the rate at which AI costs are declining is even more incredible.<ol><li>The socioeconomic value created by linear growth in intelligence is super-exponential. This result suggests that there is no reason to expect investments in exponential growth to stop in the near term.</li></ol>If these three observations continue to hold, their impact on society will be enormous.<img src="https://imageproxy.pbkrs.com/https://wpimg-wscn.awtmt.com/5cfcd32a-eb93-4192-afcf-ffb86ff2cc46.png?x-oss-process=image/auto-orient,1/interlace,1/resize,w_1440,h_1440/quality,q_95/format,jpg" alt="" original-src="https://imageproxy.pbkrs.com/https://wpimg-wscn.awtmt.com/5cfcd32a-eb93-4192-afcf-ffb86ff2cc46.png"/>We are now beginning to roll out AI Agents, which will eventually be widely used like virtual colleagues.Let’s take software engineering Agents as an example, which we believe are particularly important. Imagine that these Agents could ultimately accomplish most tasks that top company software engineers with several years of experience can complete (tasks that may take up to several days). They may not generate the most groundbreaking new ideas, require significant human oversight and guidance, and may excel in some areas while performing poorly in others.But they can still be imagined as real yet relatively junior virtual colleagues. Now imagine having 1,000 such Agents. Or 1 million. Then imagine having such Agents in every field of knowledge work.<img src="https://imageproxy.pbkrs.com/https://wpimg-wscn.awtmt.com/658105c5-f0f9-4784-9ae7-1805c99e94e4.png?x-oss-process=image/auto-orient,1/interlace,1/resize,w_1440,h_1440/quality,q_95/format,jpg" alt="" original-src="https://imageproxy.pbkrs.com/https://wpimg-wscn.awtmt.com/658105c5-f0f9-4784-9ae7-1805c99e94e4.png"/>In a sense, AI may be economically similar to transistors—a major scientific discovery that can be scaled massively and permeates every corner of the economy. We won’t focus too much on transistors or transistor companies, but we will expect our computers, televisions, cars, toys, etc., to perform wonders.The world will not change overnight; it never has. In the short term, life will remain largely the same, and people in 2025 will spend most of their time much like those in 2024. We will still fall in love, start families, argue online, hike, and so on.But the future will come at us in a way that cannot be ignored, and the long-term changes in society and the economy will be immense. We will find new things to do, new ways to help each other, and new ways to compete, but these may be very different from today’s work.Autonomy, willingness, and determination may become extremely important. Making the right decisions about what to do and how to navigate a constantly changing world will have immense value; resilience and adaptability will be skills worth cultivating. AGI will be the strongest lever of human will, enabling individuals to have a greater impact than ever before, rather than a smaller oneWe expect the impact of AGI to be uneven. While some industries may not change much, scientific progress could occur faster than today, and this impact of AGI may surpass everything else.The prices of many commodities may ultimately decline significantly (currently, the costs of intelligence and energy limit many things), while the prices of luxury goods and some inherently limited resources, such as land, may rise more dramatically.From a technological perspective, the path forward seems quite clear. However, public policy and collective opinion are crucial in how AGI is integrated into society. One reason we are launching products early is to give society and technology time to evolve together.AI will permeate all areas of the economy and society, and we expect everything to become intelligent. Many people expect that individuals should be given more control over technology than historically, including more open-source options, and accept the fact that there needs to be a trade-off between safety and personal empowerment.While we never want to act recklessly, and there may be some significant decisions and restrictions related to AGI safety that will be unpopular, we believe that as we get closer to achieving AGI, it is important to lean towards empowering individuals more. Another possible path we can see is that AI is used by authoritarian governments to control their populations through mass surveillance and deprivation of autonomy.<img src="https://imageproxy.pbkrs.com/https://wpimg-wscn.awtmt.com/f6142b5d-df72-4d36-8036-8f9f495cc447.png?x-oss-process=image/auto-orient,1/interlace,1/resize,w_1440,h_1440/quality,q_95/format,jpg" alt="" original-src="https://imageproxy.pbkrs.com/https://wpimg-wscn.awtmt.com/f6142b5d-df72-4d36-8036-8f9f495cc447.png"/>Ensuring that the benefits of AGI are widely distributed is crucial. Historical impacts of technological progress indicate that most metrics we care about (health outcomes, economic prosperity, etc.) will improve on average and in the long term, but increasing equality does not seem to be determined by technology, and achieving this correctly may require new ideas.In particular, the balance of power between capital and labor may be disrupted, which may require early intervention. We are open to some ideas that sound quite strange, such as allocating a certain &#34;computing budget&#34; to everyone on Earth so they can utilize a large amount of AI, but we can also see many ways to achieve the desired effects by continuously lowering the costs of intelligence.By 2035, anyone should be able to access intelligence capabilities equivalent to the total intelligence of everyone in 2025; everyone should be able to obtain unlimited wisdom to utilize in any way they can imagine. Today, there are many talented individuals in the world who do not have enough resources to fully showcase themselves, and if we can change this, the world will create tremendous value, bringing great benefits to all of us.Source: [AIGC Open Community](https://mp.weixin.qq.com/s?__biz=Mzg3Mzg5MjY3Nw==&amp;mid=2247517888&amp;idx=1&amp;sn=ed62122b2dff8a887fa199a7a49f4eea&amp;chksm=cf58b253da9d47502ae3b1d1c3ab552d3215136515b013d5aa7259f81e115c05365bd72c6f19&amp;mpshare=1&amp;scene=23&amp;srcid=0210AGgYtqQ4w5YdOZVPF9rP&amp;sharer_shareinfo=0c25f377ddbf035349dd0a35830861f7&amp;sharer_shareinfo_first=7bac3261b8764e95a59509cf1e4441fe#rd), Original title: &#34;Just now, Sam Altman posted late at night, AI Agent will reshape the world economy&#34;

OpenAI

80020

00020

Altman stated that the intelligence level of AI models is roughly equal to the logarithm of the resources used to train and run them; the cost of using AI at a specific level decreases by about ten times every 12 months, and lower prices lead to more usage; the socioeconomic value created by linearly growing intelligence is super-exponential

- Sam Altman, co-founder and CEO of OpenAI, published an article titled "Three Observations" discussing the economics of AI.  
- He highlights that AI model intelligence correlates with the logarithm of resources used, costs decrease significantly over time, and the socioeconomic value generated by linear intelligence growth is super-exponential.  
- Altman emphasizes the potential of AI Agents to reshape the economy, suggesting that their widespread use could lead to transformative changes across various sectors.

Just now, Sam Altman posted late at night that AI Agents will reshape the world economy