OpenAI CEO Sam Altman admits that AI’s benefits may not be widely distributed

A New articles on his personal blogOpenAI chief executive officer Sam Altman says, “To enable everyone on the world to use a lot of AI” and the benefits of technology are widely distributed, the company is open to a “count budget” among other “strange-so-sub” concepts.

“The historical influence of technical progress indicates that we are mostly careful about metric (health results, economic prosperity, etc.) average and long -term, but growing equivalent does not seem technically determined and this right is not available, new ideas are needed. “Ultman writes.” In particular, the balance of power between capital and labor can easily become rumors and it may require initial intervention. ”

It can be easier to offer than to implement the solutions to this problem, such as the concept of “count budget” of Ultman. Already, AI is affecting the labor marketResult Work And Downsizing of the departmentThe Experts have warned That is the potential result of the rise of the AI ​​Tech if the unemployment is not with the right government policies and the raising and upscilling programs.

Not the first timeUltman claimed that artificial General Intelligence (AGI) – which he defined as “[an AI] The system that can deal with growing complex problems, at the human level, in many cases ” – nearby. Whatever it takes on it, this AGI will not be perfect, Ultman warned that it is that it “may require a lot of people’s supervision and direction”

“[AGI systems] “The biggest new idea will not have,” and it will be great in some things but will be amazingly worse for others. “

However, the original value of the AGI will come from these systems to run a huge size, Ultman said the scene. OpenAI rival anthropic CEO, like Dario AmodiUltman imagines the tasks of “tackling in every field of knowledge” in thousands or even millions of hyper-enabled AI systems.

Someone can assume that it will be an expensive vision to realize. In fact, Ultman observed that “you can spend voluntary amount of money in AI performance and get continuous and estimated profit”. Openai is probably because of this Allegedly In a fund’s round to raise $ 40 billion in and contains Promised to spend up to $ 500 billion With partners on a huge data network.

Nevertheless, the Ultman also does the case that the “a layer of given AI” comes down to about 10x every 12 months. In other words, pushing the boundaries of AI technology will not be cheap, but users will get access to growing systems on the way.

From the capable, cheap AI models Chinese AI Startup DEPSEC and others seem to be Support that idea. There is Advisory Training and development expenses are also coming down, but both Ultman and Amodai have argued Invested AG-level AI-and beyond it must be achieved.

How Opeena planned to publish AGI-level systems (it can be captured, actually they can create them), Ultman said the company probably “some large decisions and limitations related to AGI protection that will be obsolete.” Open Once committed It will be committed to stop the competition as soon as any “standard-addressed,” “” protection-conscious “project comes closer to the AGI creation, as soon as the protection is out of anxiety and starting support.

Of course, when the Open was unprofitable. The agency has The process of converting the corporate structure into a more traditional, profit-powered orgThe Open Report The target and the current annual sales of Nestle are equal, the goal of reaching $ 100 billion in revenue by 2029.

It was that Ultman added that the goal of the Openai was to create a stronger AI “The tendency would be increased” when “AI” was being used to control their population and to control their population through autonomy. ” Ultman recently said that he thought that Open was on the wrong side of history when its technologies came in the open-sourcing. Open is in the past, Open-tacked technology, the company is usually a owned, in favor of the approach to the development of closed-source.

“AI will enter the economy and society in all areas; We hope everything is smarter, “Ultman said.” Many of us hope to give us more control over technology, including Open-Sorcing to us, and it acknowledges that there is a balance between security and individual empowerment for which trade- Will be required of off. “

Ultman’s blog post arrived in Paris before this week’s AI Action Summit, which has already encouraged other technologies to outline notable Their own View For the future of AI.

In a footnote, Ultman has added that OpenAI actually does not plan to finish his relationship with close partner and investor Microsoft at any time using the word AGI. Microsoft and Open Reported that the AGI had a contractual definition of AGI – AI systems that can earn $ 100 billion profit – which allows Open to discuss the conditions of more favorable investment once in combination. Ultman says, however, Open A. “fully expects[s] Long -term partnership with Microsoft. “