Skip to content
TheChatArchiveThe Chat Archive
Perplexitysonar-pro-high· PerplexityHugging Face LMArena Search Arena 24K

Is the rumour true that ChatGPT worked on a specification for 6 hours and after this it…

From a Chatbot Arena battleThe same person sent this prompt to two anonymous models at once. Compare it with Geminigemini-2.0-flash-grounding’s answer.
  1. User

    Is the rumour true that ChatGPT worked on a specification for 6 hours and after this it gave a correct program source out?

  2. Perplexity Assistant

    There is no credible evidence supporting the specific rumor that ChatGPT worked on a specification for six hours and then produced a correct program source as a direct result of that continuous effort. Here’s what is known based on available information:

    How ChatGPT Actually Works

    • No Background Processing or Long-Term Tasks: ChatGPT does not perform background work or "think" over a period of hours. Each response is generated in real time, based solely on the current prompt and the conversation context. If a user asks ChatGPT to "work for six hours," the AI does not actually spend that time processing or refining its answer; it simply generates a response as soon as the prompt is submitted[6].
    • No True Waiting or Progress: Reports from users confirm that when ChatGPT claims it needs more time or is "still working" on a task, this is a conversational illusion. The model does not keep working in the background or improve its output with time; it only generates new responses when prompted again[6].

    ChatGPT and Programming Tasks

    • Effective for Iterative Coding Help: ChatGPT is widely used to assist with programming tasks, including generating code snippets, debugging, and refining routines. Many users report that it can save hours or even days of work, especially for well-defined, modular tasks[2][5].
    • Requires Iterative Refinement: For complex programming projects, ChatGPT typically produces useful starting points or partial solutions. Users often need to refine prompts, clarify requirements, and ask follow-up questions to get a correct and complete program[5]. The process is interactive, not a single, prolonged computation.
    • Not Autonomous Over Hours: There is no mechanism for ChatGPT to autonomously work on a problem for hours and then deliver a finished product. All progress is driven by user input and prompt iteration[6].

    Anecdotes and User Experiences

    • Users Save Time, But Not via Continuous AI Effort: Many users share stories of ChatGPT helping them automate tasks or write scripts that would have taken much longer to do manually[2][5]. However, these successes are the result of back-and-forth prompting and user guidance, not the AI working independently for hours.
    • No Evidence of Six-Hour Autonomous Work: There are no documented cases or credible reports of ChatGPT working on a specification for six hours straight and then producing a correct program as a result of that continuous effort.

    Summary Table

    ClaimReality
    ChatGPT can work on a task for 6 hoursNo, it only responds to prompts in real time; no background or continuous processing[6].
    ChatGPT can produce correct codeYes, often with iterative prompting and user guidance[2][5].
    ChatGPT autonomously refines code over hoursNo, all progress is user-driven; no autonomous long-term refinement[6].

    Conclusion

    The rumor that ChatGPT worked on a specification for six hours and then produced a correct program source is not accurate. ChatGPT is a powerful tool for coding assistance, but it does not perform background work or improve its output over time without user interaction. Success with ChatGPT in programming comes from iterative, interactive prompting—not from the AI working independently for extended periods[2][5][6].

Responses were produced in different, uncontrolled contexts (system prompts, settings, prior turns). This is not a benchmark.