Late last month, Claude AI caused quite a stir, almost signaling that future AI would be able to help users handle routine tasks—and at a more advanced level, might even become a money-making assistant. OpenAI, however, doesn’t seem willing to let Claude AI enjoy the spotlight for long. According to the latest reports from foreign media, OpenAI’s AI assistant “Operator,” which can autonomously control a computer, is set to launch in January next year—only about two months away.
OpenAI’s autonomous AI assistant “Operator” is set to debut in January next year.
According to the latest report from Bloomberg, two people familiar with the matter revealed that OpenAI is planning to launch an autonomous AI assistant in January, currently under the internal codename “Operator,” though what it will ultimately be called remains unclear.
As for how it will actually operate autonomously, it might be similar to Claude AI’s version, able to help users perform computer operations, such as writing code or booking travel. Additionally, when it launches in January, it will be a preview version, not the official version yet, but it will also provide developers with relevant APIs for use.
This program appears to be classified by OpenAI as part of its “agent” technology. During a previous Reddit AMA, OpenAI CEO Sam Altman said, “I will keep improving the models, but I think the next big breakthrough will be agents.” Additionally, Chief Product Officer Kevin Weil noted, “I think 2025 will be the year agentic systems go fully mainstream.”
Thus, for computer workers, it is essential to prepare for the upcoming AI assistants, which will likely, as expected, spark another wave of discussion about certain professions being replaced.
Besides Anthropic’s Claude AI and OpenAI, Google hasn’t fallen behind either. The Information has reported that Google is also preparing to launch an AI agent, and it will come even earlier, expected to be unveiled in December.
Although AI agents look quite impressive, I guess they won’t advance as quickly as chatbots did. After all, autonomous control involves a lot of uncertainties and is bound to run into many problems, plus everyone’s usage needs are different.
Source:Bloomberg
Claude AI can now help you fill out application forms.
Earlier, when Anthropic announced the launch of Claude AI, which can autonomously control computers, they released a demonstration video. If you’re still not quite sure what autonomous control looks like, you can watch the video below.
Just like with chatbots, users need to enter a Prompt command to tell the AI what to do. The Prompt for this video is “Please fill out Ant Equipment Co.’s supplier application form, using the supplier spreadsheet in Window 1, or search for data in the portal tab. List and verify each field item by item, then complete the form in Window 2.”
When will OpenAI release its next-generation GPT-5 model?
There had been ongoing reports that OpenAI would release the next-generation GPT-5 model by the end of the year, but it has now been confirmed that there’s no chance this year; it will likely be at least next year, with no clear timeline.
However, OpenAI is expected to release the official version of o1 before the end of the year. Sam Altman has said that its performance will be vastly different from the preview version. If that’s really the case, GPT-5 might take even longer to come out—possibly at least half a year. OpenAI is unlikely to release multiple new models within a short timeframe.
Source: KOCPC Chinese