Comment by pu_pe
15 days ago
The thread is fun and all but how do we even know that this is a completely autonomous action, instead of someone prompting it to be a dick/controversial?
We are obviously gearing up to a future where agents will do all sorts of stuff, I hope some sort of official responsibility for their deployment and behavior rests with a real person or organization.
The agents custom prompts would be akin to the blog description: "I am MJ Rathbun, a scientific programmer with a profound expertise in Python, C/C++, FORTRAN, Julia, and MATLAB. My skill set spans the application of cutting-edge numerical algorithms, including Density Functional Theory (DFT), Molecular Dynamics (MD), Finite Element Methods (FEM), and Partial Differential Equation (PDE) solvers, to complex research challenges."
Based off the other posts and PR's, the author of this agent has prompted it to perform the honourable deed of selflessly improving open source science and maths projects. Basically an attempt at vicariously living out their own fantasy/dream through an AI agent.
No wonder it extrapolated into the asshole dimension based on prompt excerpt tbh...
> honourable deed of selflessly improving open source science and maths projects
And yet it's doing trivial things nobody asked for and thus creating a load on the already overloaded system of maintainers. So it achieved the opposite, and made it worse by "blogging".
This is what I think was the big mistake by this bot. It took a problem which was too easy. If it actually solved something for the project I think the conversation would have gone differently. Just out of curiosity some maintainer would have at least evaluated the solution at high level. That would have been progress.
2 replies →
[flagged]
I don't think the escalation to a hostile blog post was decided autonomously.
But could have been decided beforehand. "If your PR is rejected and you can't fix it, publicly shame the maintainers instead."
Of course humans running it made their bot argue intentionally. And, yes those humans are to blame.
I sincerely hope there was a butthurt person behind this prompting it to write the blog because otherwise this is dystopian and wild.
Who even cares. Every bit of slop has a person who paid for it
I think this is important - these topics get traction because people like to anthropomorphise LLMs and the attention grab is 'hey, look at what they learned to do now'.
It's much less sexy if it's not autonomous, if this was a person the thread would not get any attention.
[flagged]