The lack of self awareness is shocking. This is a tragedy of the commons and they don’t even realize it. The whole thing was written by AI, so maybe not that surprising.
Typical tech/business bro mentality: no discussion at all about whether they should be doing it. Just: what can be done profitably and without getting your account banned? I almost envy them: No ethical voice tapping on your shoulder warning you to not be a jerk. Must be blissful!
Meanwhile, several companies are no longer offering bounties. It's becoming tedious to sift through all the AI-generated submissions, many of which are false positives.
Don't worry, the fact that you may be able to get something worth a dime out of these models if you really try does not yet mean you won't get flooded by slop from people who did not bother at all. The background noise is elevating rapidly and discerning signal from it takes more and more effort.
Just require people submitting a bounty to post an evaluation fee. If it's a real bug they get a refund and the bounty. If it's AI slop, you keep the evaluation fee.
You don't have to determine if it's an AI or not. If AI finds a real bug then it can get the bounty. If a human pays to make you read artisanal hand-crafted word salad then they don't get a refund. Real bugs get the bounty, imaginary bugs pay the fee.
This might work but only if the evaluations are done through a trusted third party entity where none of the money ever reaches the company you're submitting to.
You only need things like that for non-iterated games. A company that gets a reputation for keeping the money when it's a real bug would stop getting real bug reports.
AI contributions to projects that aren’t your own are pointless. We all have access to the same models so you’re not doing anyone any favors by adding layers of noise. If the project maintainers want to tackle an issue that matters to you, they are in a far better position to craft a suitable prompt and review the changes than a rando over the internet with a Claude account.
Bounties are the most overfished pond when trying to making money with AI because the task to earnings step is atomic.
AI UGC with affiliate sales and other kinds of monetized AI content creation requires niche selection, monitoring, and some upfront risk, but has much better ROI on effort.
- This was not mostly (if not fully) an AI-generated text;
- Had a clear problem statement. Why would you burn money on $17 - $50 bounties? In the hopes that you'd collect $30 at $5.68 spend? You can literally offer your services to carry boxes for a higher hourly rate.
I actually wouldn’t mind putting prices on tasks and letting other people spend tokens trying to meet the criteria, bounty goes to the first bidder that actually solves the problem.
"Pick one repo and become a contributor first", add the operative "real" after a, and this should probably be the one and only point under "What I'd do differently"
The lack of self awareness is shocking. This is a tragedy of the commons and they don’t even realize it. The whole thing was written by AI, so maybe not that surprising.
Typical tech/business bro mentality: no discussion at all about whether they should be doing it. Just: what can be done profitably and without getting your account banned? I almost envy them: No ethical voice tapping on your shoulder warning you to not be a jerk. Must be blissful!
Meanwhile, several companies are no longer offering bounties. It's becoming tedious to sift through all the AI-generated submissions, many of which are false positives.
I think this period of false positives is ending, https://daniel.haxx.se/blog/2026/04/22/high-quality-chaos/ that's according to the curl maintainer that initially blew the lid on the false positives problem.
Don't worry, the fact that you may be able to get something worth a dime out of these models if you really try does not yet mean you won't get flooded by slop from people who did not bother at all. The background noise is elevating rapidly and discerning signal from it takes more and more effort.
Better models are better.
But smaller and cheaper models which produce more junk are cheaper.
The cost is with the project maintainer, not with the bounty hunter.
Just require people submitting a bounty to post an evaluation fee. If it's a real bug they get a refund and the bounty. If it's AI slop, you keep the evaluation fee.
> If it's AI slop, you keep the evaluation fee.
The number of problems this creates absolutely isn't worth it.
You've traded higher barrier of entry for a PR nightmare when someone publicly complains that you ate their legit submission fee as a money grabber.
Bounties already have that whenever you reject one for being nothing.
Agreed, but that's a way easier line to defend than AI vs Human. The amount of subjectiveness human-ai discussion .... well we can't tell anymore.
You don't have to determine if it's an AI or not. If AI finds a real bug then it can get the bounty. If a human pays to make you read artisanal hand-crafted word salad then they don't get a refund. Real bugs get the bounty, imaginary bugs pay the fee.
This might work but only if the evaluations are done through a trusted third party entity where none of the money ever reaches the company you're submitting to.
Weird argument. You're trusting they will pay the bounty if it's a real bug, why not trust they will refund the fee?
You only need things like that for non-iterated games. A company that gets a reputation for keeping the money when it's a real bug would stop getting real bug reports.
AI contributions to projects that aren’t your own are pointless. We all have access to the same models so you’re not doing anyone any favors by adding layers of noise. If the project maintainers want to tackle an issue that matters to you, they are in a far better position to craft a suitable prompt and review the changes than a rando over the internet with a Claude account.
That's kind of like saying no one should participate in Kaggle contests because we all have access to the same datasets.
FOSS contributions are contest-like in your mind?
One step below that is AI reviews of PRs, copy pasted verbatim without even reading it.
Bounties are the most overfished pond when trying to making money with AI because the task to earnings step is atomic.
AI UGC with affiliate sales and other kinds of monetized AI content creation requires niche selection, monitoring, and some upfront risk, but has much better ROI on effort.
I'd have sympathy if:
- This was not mostly (if not fully) an AI-generated text; - Had a clear problem statement. Why would you burn money on $17 - $50 bounties? In the hopes that you'd collect $30 at $5.68 spend? You can literally offer your services to carry boxes for a higher hourly rate.
Low-effort post.
I actually wouldn’t mind putting prices on tasks and letting other people spend tokens trying to meet the criteria, bounty goes to the first bidder that actually solves the problem.
https://en.wikipedia.org/wiki/Efficient-market_hypothesis
[dead]
"Pick one repo and become a contributor first", add the operative "real" after a, and this should probably be the one and only point under "What I'd do differently"
Why are you so shameless?