Can't believe I missed this one -- it has been up for a while now: For those who like to take these yearly surveys; the 2025 Developer Survey | Stack Overflow is up.
Be prepared to answer a lot of questions about AI/LLM's.
Can't believe I missed this one -- it has been up for a while now: For those who like to take these yearly surveys; the 2025 Developer Survey | Stack Overflow is up.
Be prepared to answer a lot of questions about AI/LLM's.
This one got so annoying and repetitive that at one point I started regretting filling it out this year. It must have been half the survey, which felt longer to finish (probably because it was such a joyless endeavour) than the years prior.
Thanks for the warning. I'll skip this one then, thank you very much!
With the platform itself growing steadily irrelevant by the day, can't help but wonder how meaningful the results are going to be. How many of the people who (still) use SO on a regular basis will be keen on sitting through the whole thing? How well the preferences / answers / biases of those will match those of everybody else: including the folks who'd much rather keep on copy-pasting/vibe-coding/agent-izing via their favourite <X>GPT of the day than to complete this?
Indeed. I used to value SO's surveys, but I'm not sure they're on the right track any more.
I saw someone on Stack Exchange quip "It looks like this survey was generated by an AI that has a vested interested in its own self-preservation".
I eventually finished it, but it was just out of spite and sunk-cost fallacy. Had I known how bad it would be, I wouldn't have bothered.
Same. I completed it but, had I known the questions in advance, I wouldn't have taken the time. It might still be interesting to see the outcome related to programming languages and frameworks, though.
Either I forgot or missed that item last year, but it's as I suspected: the partnership with OpenAI
must be one good motive for the biased content.
I no longer fill this thing out because I feel that SO users do not reflect a sufficiently diverse or generic programming population so as to be able to infer anything useful from the aggregate responses.
I took the survey. It seemed very biased towards AI and LLMs. It was rather disappointing.
And it took me around 30 minutes to complete.
They’ve published their results, if anyone’s interested.
I’m taking those with a grain of salt, given the turn the SO survey has taken, but it was interesting to see a new Gleam language in 2nd position as “most desired” languages, out of the blue (and most of its sources are currently in Rust
).
That category would indeed be more relevant if people desiring the language had actually programmed with it already, even if not intensively, which isn’t a requirement in the survey (desired: not used this year, but want to use next year / admired: used this year and want to use next year. Although this isn’t entirely clear given the results in Programming, scripting, and markup languages, which don’t match the admired percentages.)
Positive sentiment to AI tools has decreased in 2025
More developers actively distrust the accuracy of AI tools than trust it
The biggest single frustration, cited by 66% of developers, is dealing with "AI solutions that are almost right, but not quite," which often leads to the second-biggest frustration: "Debugging AI-generated code is more time-consuming" (45%)
Gives me some hope this AI bubble will burst.
Rust related:
Cargo is the most admired cloud development and infrastructure tool this year
Cloud, Cargo? Did I miss something? ![]()
uv is the most admired SO tag technology this year
Spreading the joy of cargo to other languages, written in rust ![]()
One in four developers are happy at their current job
A glass quarter full observation.
This topic was automatically closed 90 days after the last reply. We invite you to open a new topic if you have further questions or comments.