AI agents becoming a commercial "lifeline"? Microsoft, Google, and OpenAI are actively preparing
AI agent technology is gradually becoming a key tool for enterprise automation, capable of automatically completing various complex tasks such as shopping and travel planning. Tech giants like Microsoft, Google, and OpenAI are competing to develop their own versions of AI agents to earn more revenue. Microsoft is developing intelligent assistants to help salespeople automatically process orders and invoices, and enhance their capabilities to handle more complex tasks. In addition, Microsoft also provides a tool for software developers that can automatically provide solutions and help write and run code. This trend is expected to become an essential tool for enterprises, improving work efficiency
Source: Hard AI
Author: Fang Jiayao
Tech giants are competing to invest in "AI agents" software, aiming to develop this feature into an essential tool for enterprises to earn more revenue.
Imagine having a very smart intelligent assistant on your computer. While you are busy working, it can help you plan your holiday trips, including booking flights and hotels. Isn't that convenient?
AI agent software is an efficient intelligent tool that can automatically handle various tasks such as email management, travel scheduling, and even assist in programming. This allows people to focus more on other important tasks at work.
Tech giants like Microsoft, OpenAI, and Google are racing to develop their own versions of AI agents. If AI agents can truly help companies save time and money, these tech giants can earn more revenue by selling AI agent technology.
Tech Giants Competing to Develop Their Own AI Agent Versions
Taking Microsoft as an example, according to media reports, the company is developing an "AI agent" intelligent assistant that can help salespeople automate order processing and invoicing. In the past, salespeople had to manually complete multiple steps to process orders and issue invoices, but now, this intelligent assistant can automatically handle these tasks. Specifically, if the system detects a large order that has not been processed, the intelligent assistant will automatically create an invoice and ask the salesperson if it should be sent to the customer. Once the invoice is sent, the intelligent assistant will also track the customer's response and payment status, automatically saving this information to the company's system.
Furthermore, Microsoft is enhancing the capabilities of these intelligent assistants to handle more complex tasks, such as writing and running code. GitHub Copilot is a tool provided by Microsoft for software developers. Developers only need to describe the programming problems they encounter to this tool, and the intelligent assistant will automatically provide solutions, even helping to write and run new code.
It is reported that Microsoft plans to showcase some features of AI agents at the upcoming annual Build developer conference next month.
Reportedly, companies like OpenAI, Google, and Meta are all competing to develop more intelligent and workflow-simplifying advanced computer assistants.
For example, OpenAI is secretly developing a computer-use agent that can operate multiple applications simultaneously, such as helping users transfer data from documents to spreadsheets. In addition, OpenAI is also collaborating with Meta to develop AI agents that can handle more complex network tasks, such as automatically planning travel routes and booking hotels.
At the same time, the startup company Adept is developing a computer-use agent. According to sources familiar with Adept's plans, the company has raised over $400 million and plans to launch the product in the summer. This AI assistant learns operations by watching videos of people working on computers, such as creating Excel spreadsheets, enabling it to perform tasks on the computer like a human, such as automatically searching for online housing listings or recording customer calls
Companies expect AI agent technology to bring substantial changes
With the popularization of the concept of AI agents, companies need to clearly define their functions to avoid customer confusion.
Originally, the term "AI agent" was used to describe intelligent programs that can autonomously complete tasks. However, with technological advancements, many companies have started using "AI agent" to describe various intelligent tools, sometimes these tools are not as "intelligent" as initially expected.
For example, Google recently announced its so-called "AI agents," which are essentially different versions of the chatbot ChatGPT, mainly used for specific occasions such as assisting in customer service chat to solve problems. However, these bots cannot handle a variety of complex tasks autonomously like the original AI agents.
Currently, the term "AI agent" is being used too broadly, to the point where people are not clear about what an AI agent can actually do. If a company claims that their "AI agent" is powerful, but in reality, this agent can only work within a very narrow range, users' high expectations often turn into disappointment.
Moreover, companies expect AI agent technology to bring substantial changes. MongoDB's CEO Dev Ittycheria also pointed out that although some AI features have started to be integrated into applications like email, these features have not fundamentally changed his work life.
Ittycheria and his team are looking forward to more powerful AI capabilities, as they believe that "AI agents" will be the next major breakthrough in AI technology