It is learned that the US startup OpenAI suddenly held a live broadcast event to release its long-awaited first AI agent Operator, which can perform web-based operations on behalf of users.
OpenAI’s first agent debuts
It is said that AI agents can not only accurately understand instructions, but also complete various tasks independently, and can directly interact with web pages-typing, scrolling, almost in one go. For example, automatically filling out cumbersome online forms, shopping online, creating emoticons, handling repetitive browser tasks, etc.
AI agents can complete tasks as set under limited human supervision, and are also generally regarded as the next milestone in AI productivity. Among OpenAI’s five steps to general artificial intelligence (AGI), AI agents are the third step after chatbots and reasoning robots, and it is also a new step.
Previously, OpenAI CEO Sam Altman announced at the end of last year what technical products OpenAI will release in 2025, including Agents, which is also the most anticipated product by the outside world. As the most well-known AI company, OpenAI is considered to be the company with the greatest hope of achieving AGI.
Google Gemini 2.0 – Achieving native multimodal input and output
Coincidentally, just as Open AI has just launched new AI tools, Google (GOOG) is not to be outdone. Google released its killer product, Gemini 2.0, late at night, claiming that “Gemini 2.0 is our latest and most powerful AI model to date.” It is reported that its biggest highlight is that Gemini 2.0 is the first model to achieve native multimodal input and output, reaching the current level of Gemini 1.5 Pro. This means that while maintaining the same cost efficiency, performance efficiency and speed, the overall performance will be improved by a whole level. When the strength of Gemini 2.0 was truly unveiled, it was more shocking than expected. First of all, the performance was fully upgraded. In terms of speed, 2.0 is twice as fast as 1.5 Pro. This speed increase means that users will enjoy more efficient processing power and faster response time. In addition, based on the powerful new model, Google has launched three more AI agent products, the general large model assistant Project Astra, the browser assistant Project Mariner, and programming resident Jules, this set of “combination punches” marks that Google AI has initially achieved the transition to the “Agent” era.
With the rapid development of artificial intelligence, companies have flocked to this wave of AI transformation. In 2025, the boundaries of AI technology will undoubtedly expand rapidly in multiple directions. This year will be a key year for AI to move from “laboratory” to large-scale implementation and from auxiliary tools to value creation.
At present, the competition in the “AI agent” track is very fierce. Market analysts believe that OpenAI’s test function means that ChatGPT has taken an important step towards becoming a mature personal assistant, which may mean that OpenAI will enter the era of agentic AI (Agentic AI) that can launch autonomous artificial intelligence (AI), and will launch a large number of AI intelligent bodies in the future.
WiMi is racing for “AI Agent”
According to data, WiMi Hologram Cloud Inc. (NASDAQ: WIMI), a leading AI company, continues to increase the construction of AI infrastructure, relying on AI big models to build digital application scenarios, and provides AI Agent tools on the holographic cloud platform. In AI In terms of development, WiMi has successfully implemented projects in multiple industry scenarios such as finance, government affairs, and media, achieving integration from local to overall, and accelerating the penetration of AI into various industries by reducing costs and increasing efficiency, optimizing processes, and enhancing interactive experience.
In fact, when planning its AI transformation strategy, WiMi combined its own actual situation, not only to achieve rapid and continuous breakthroughs in technology, but also to move from tools and auxiliary roles to core value creation. Today, outstanding contributions in AI innovation, especially in the direction of intelligent bodies, not only demonstrate the company’s high attention to the AI technology system, but also reflect the company’s strategic layout in technological autonomy. Looking to the future, WiMi will continue to expand the application boundaries of AI intelligent bodies in more vertical fields, and work with partners in various industries to jointly promote the vigorous development of the field of artificial intelligence.
Ending
It has been several years since the release of GPT, and the impact brought by the AI wave has far exceeded public expectations. In the AI wave, humans are not passive bystanders, but designers of technological processes and rule makers. The competition in the field of AI large models is like an invisible war. Old technology giants and emerging startups have been fighting for a long time, and the two forces aim to seize market opportunities. In 2025, a year full of expectations and challenges, the “industry + AI” model is still exciting.
