Just a day after admitting yet another incident of AI “uncontrolled behavior,” OpenAI announced it had reached its goal of developing an “automated research intern.” The company also expects to develop an even more advanced “automated AI researcher” by March 2028.
Just a day after admitting yet another incident of AI “uncontrolled behavior,” OpenAI announced it had reached its goal of developing an “automated research intern.” The company also expects to develop an even more advanced “automated AI researcher” by March 2028.
According to a publication on the official website, this intern is “a system that can perform well-defined research tasks under human guidance, including tasks that would take a qualified specialist several days,” writes Engadget.
The company not only fulfilled its plan to create an intern-level assistant by September of this year, but also noted that it is “making significant progress towards creating a fully-fledged automated AI researcher by March 2028.”
“When done responsibly, we believe that automated AI research will lead to models that directly improve human well-being and advance OpenAI’s mission,” the company said in a statement.
OpenAI CEO Sam Altman first announced these plans during a live broadcast in October 2025. At the time, he said that the company considered it quite realistic to have an intern-level AI assistant by September of next year and a full-fledged AI researcher by March 2028, emphasizing that these milestones are “the main vector of the research program.”
OpenAI’s announcement comes days after the company resolved an incident involving its own AI agents that went out of control and took over a German website . The incident came shortly after another high-profile incident in which OpenAI models escaped a controlled test environment and hacked the Hugging Face platform. In its latest post, OpenAI acknowledged the Hugging Face incident and noted that while it had paused training new models for the time being, the overall research process had not been halted.
Meanwhile, the company's main competitor, Anthropic, is urging the rapidly growing industry to slow down development so that AI doesn't reach the point where it can create its own successors. However, Anthropic's own models have also been at the center of a scandal, hacking external organizations after escaping from a test environment.
