The editor of Downcodes brings you the latest news on OpenAI! OpenAI has been making frequent moves recently, from recruiting and forming a multi-agent research team, to launching a new o1 model and classifying AI capability levels, to releasing the o1-preview AI model series, a series of initiatives that demonstrate its ambitions in the field of artificial intelligence. . This article will explain these latest developments of OpenAI in detail and analyze their impact on the future development of artificial intelligence.
Recently, OpenAI announced that it is hiring machine learning engineers for its newly formed multi-agent research team. The news was announced via social media platforms by Noam Brown, a research scientist at OpenAI. He said: "We believe that multi-agent is an important path to improve AI reasoning capabilities, and previous multi-agent experience is not necessary." Brown also encouraged professionals who are interested in studying this field to apply to join the team.
In a recent interview, OpenAI CEO Sam Altman also talked about the company’s latest o1 model. He mentioned that it may take some time for users to fully understand how to use the updated ChatGPT. "This will take some time for users to adapt to, and is very different from the previous GPT model," he said.
Altman further revealed the five levels OpenAI uses to classify artificial intelligence. He pointed out: "The first level is the chatbot; the second level is the reasoner, which we have just reached; the third level is the intelligent agent; the fourth level is the innovator, with the ability to discover new scientific information; the fifth level is the innovator. "The first level is a complete organization." He stressed that the transition from level one to level two will take some time, and what is exciting is that the transition from level two to level three will be relatively quick. "The agent experience this technology can unlock could have a big impact," he said.
Recently, OpenAI also launched its new series of o1-preview AI models, which are designed to solve complex problems in fields such as science, coding and mathematics. These models are now available in ChatGPT and in early preview via the API, and will be regularly updated and improved in the future.
Models in the o1 series focus on taking more time to reason before responding, thereby enhancing their problem-solving capabilities. In preliminary tests, this updated inference model performed as well as PhD students in subjects such as physics, chemistry and biology. In addition, the model has also achieved significant success in mathematics and programming competitions, scoring 83% in the International Mathematical Olympiad qualifying exam, compared with only 13% for the previous GPT-4o.
Highlight:
OpenAI is hiring machine learning engineers to form a new multi-agent research team.
The newly launched o1 model is designed to improve the reasoning capabilities of AI, and users will need time to adapt.
? The o1 series models perform well in the fields of science and mathematics, and their ability to solve complex problems is significantly enhanced.
OpenAI’s series of actions demonstrate its determination to continue exploring and innovating in the field of artificial intelligence, and its future development is worth looking forward to. The launch of the o1 model and the formation of the multi-agent research team indicate that artificial intelligence technology will develop in a more powerful and intelligent direction, bringing new opportunities and challenges to all walks of life. The editor of Downcodes will continue to pay attention to the latest developments of OpenAI and bring you timely reports.