On Thursday, OpenAI introduced that ChatGPT customers can now department conversations into a number of parallel threads, serving as a helpful reminder that AI chatbots aren’t individuals with mounted viewpoints however moderately malleable instruments you possibly can rewind and redirect. The corporate launched the characteristic for all logged-in internet customers following years of person requests for the aptitude.
The characteristic works by letting customers hover over any message in a ChatGPT dialog, click on “Extra actions,” and choose “Department in new chat.” This creates a brand new dialog thread that features all of the dialog historical past as much as that particular level, whereas preserving the unique dialog intact.
Consider it virtually like creating a brand new copy of a “doc” to edit whereas maintaining the unique model secure—besides that “doc” is an ongoing AI dialog with all its amassed context. For instance, a advertising staff brainstorming advert copy can now create separate branches to check a proper tone, a humorous method, or a completely completely different technique—all stemming from the identical preliminary setup.
The characteristic addresses a longstanding limitation within the AI mannequin the place ChatGPT customers who wished to attempt completely different approaches needed to both overwrite their current dialog after a sure level by altering a earlier immediate or begin utterly contemporary. Branching permits exploring what-if situations simply—and in contrast to in a human dialog, you possibly can attempt a number of completely different approaches.
A 2024 research carried out by researchers from Tsinghua College and Beijing Institute of Know-how prompt that linear dialogue interfaces for LLMs poorly serve situations involving “a number of layers, and lots of subtasks—comparable to brainstorming, structured data studying, and huge undertaking evaluation.” The research discovered that linear interplay forces customers to “repeatedly evaluate, modify, and replica earlier content material,” growing cognitive load and decreasing effectivity.
Some software program builders have already responded positively to the replace, with some evaluating the characteristic to Git, the model management system that lets programmers create separate branches of code to check adjustments with out affecting the primary codebase. The comparability is smart: Each can help you experiment with completely different approaches whereas preserving your unique work.





