Intelligent collaboration: ChatGPT supports without replacing creativity and development

Valve engineer Fletcher Dunn shared that ChatGPT was used to optimize the development of Deadlock. In a recent post on X, Dunn explained that he described an imaginary algorithm for the game to ChatGPT and received the suggestion to apply the Hungarian algorithm. This suggestion was successfully integrated into the game’s matchmaking system.

The strength of ChatGPT compared to a typical Google search is its ability to provide accurate answers even from vague descriptions. The engineer highlighted how this approach is particularly useful when the exact technical terms needed for a traditional search are unknown.

Here’s the question:

Question:

Is there a bipartite matching algorithm where one side expresses preferences in terms of a score? In the stable marriages algorithm, both sides have a preference list. I want an algorithm where only one side has any preferences, but they are expressed as a score, and some preferences may have the same score. The best matching would be the one that minimizes the total score across all matches.

Answer:

Yes, the problem you’re describing is a variant of a weighted bipartite matching problem, where one side expresses preferences as scores, and the goal is to minimize the total score across all matches. This is known as the minimum weight matching problem on a bipartite graph, and it can be solved using algorithms like the Hungarian algorithm (also known as the Kuhn-Munkres algorithm).

A recent article in The New Yorker analyzes the use of ChatGPT as a conversational partner and a tool to support structure and develop ideas. For many people, using ChatGPT during the writing process is seen as a new form of creative collaboration rather than cheating. In fact, it is often perceived more as a sounding board to explore ideas rather than a tool to produce ready-to-use texts, thus silencing the plethora of doubts about originality and plagiarism. In practice, AI doesn’t provide shortcuts and doesn’t write on behalf of the writer, but simply facilitates certain parts of the creative process.

For all its inefficiencies, this indirect approach did feel easier than staring at a blank page; “talking” to the chatbot about the article was more fun than toiling in quiet isolation. In the long run, I wasn’t saving time: I still needed to look up facts and write sentences in my own voice. But my exchanges seemed to reduce the maximum mental effort demanded of me. Old-fashioned writing requires bursts of focus that remind me of the sharp spikes of an electrocardiogram. Working with ChatGPT mellowed the experience, rounding those spikes into the smooth curves of a sine wave.

Originally published at Levysoft.