Skip to content
Open navigation
OpenOpen workspace

Stack Overflow Developer Survey 2026

AI mistakes that cost time or money by age

In the 25-34 years old group, 62% selected “Hard-to-review generated code”; in the 65 years or older group, 26% did. Of the 10 answers tested, “Hard-to-review generated code” differs most by age, a small association: χ²(7, N = 13648) = 365.99, Holm-adjusted p < .001, V = .16. Respondents could select more than one answer, so the percentages add to more than 100%. These are unweighted respondents, not population estimates.

Published by crosstabs.com · Last updated ·
How results are checkedNamed statistical review pending. Selected automated reference cases are published on the validation page.

The table

The table

Question asked: What kinds of AI mistakes have cost you or your team the most time in the past year? Select all that apply. Respondents could select more than one answer. Shown by age, as the percentage of each group's respondents who selected each answer, with the number of respondents beneath. A respondent is counted in every row they selected, so a column adds to more than 100%.

AI mistakes that cost time or money by age: percentage of each group's respondents who selected each answer, and respondent counts
AI mistakes that cost time or moneyUnder 18 years old (A)18-24 years old (B)25-34 years old (C)35-44 years old (D)45-54 years old (E)55-64 years old (F)65 years or older (G)Prefer not to say (H)All
Plausible but wrong code69.4%G24369.0%G1,18871.0%G2,78269.3%G2,80368.8%G1,45868.8%G67756.6%23368.0%6869.3%9,452
Hard-to-review generated code48.3%FG16953.6%FG92361.5%ABDEFG2,41057.3%AEFG2,31949.7%FG1,05337.7%G37125.7%10652.0%G5254.2%7,403
Missing context48.6%FG17053.2%EFG91654.6%DEFGh2,13950.6%EFG2,04842.7%G90438.6%38031.1%12839.0%3949.3%6,724
Hallucinated package/config52.0%FG18246.8%FG80647.1%FG1,84345.4%Gf1,83644.7%G94740.0%G39425.7%10647.0%G4745.1%6,161
Incorrect or stale internal knowledge52.3%BCDEFG18342.5%DEFG73241.8%DEFG1,63938.0%FG1,53634.4%72931.0%30529.4%12142.0%4238.7%5,287
Outdated APIs42.0%DEFG14737.9%DFGe65235.7%FG1,40032.9%FG1,33233.2%FG70325.6%25224.3%10035.0%3533.9%4,621
Poor debugging path31.4%EFGd11030.2%DEFG52028.2%DEFG1,10524.2%EFG97820.1%G42616.0%15712.9%5328.0%Gf2824.7%3,377
Security/privacy issue26.3%FG9227.4%EFG47225.9%EFG1,01423.8%EFG96320.3%G43016.1%G1589.5%3937.0%EFGd3723.5%3,205
Licensing/source uncertainty18.0%DEFGc6313.4%EFG23112.4%FGe48710.9%G4429.9%2098.6%855.6%2320.0%EFG2011.4%1,560
Other, please specify11.4%407.5%1297.4%2908.1%3269.6%c2049.8%9614.8%BCDE6126.0%ABCDEF268.6%1,172
Respondents3501,7223,9174,0442,11998441210013,648

A letter beside a percentage names the columns it is significantly higher than (uppercase 95%, lowercase 90%; pairwise column tests, Bonferroni-corrected). 17,255 of 30,903 respondents have no answer to one of the two questions (not asked or not answered) and are left out.

Analysis

Results

Chi-square test of “Hard-to-review generated code”

χ² = 365.99

df = 7, Holm-adjusted p < .001

Effect size

Cramér's V = 0.164

a small association

The strongest answer, APA-style: χ²(7, N = 13648) = 365.99, Holm-adjusted p < .001, V = .16. Each answer is tested on its own, as the respondents who selected it against those who did not, across the 8 groups (N = 13,648). The p-values are Holm-adjusted for the 10 answers tested; 10 of them differ by age at the 0.05 level.

AI mistakes that cost time or money by age: the test of each answer, selected against not selected, by age
AnswerSelectedχ² (df = 7)Adjusted pCramér's V
Plausible but wrong code69.3%37.41< .0010.052
Hard-to-review generated code54.2%365.99< .0010.164
Missing context49.3%198.91< .0010.121
Hallucinated package/config45.1%87.81< .0010.080
Incorrect or stale internal knowledge38.7%111.59< .0010.090
Outdated APIs33.9%77.73< .0010.075
Poor debugging path24.7%158.97< .0010.108
Security/privacy issue23.5%126.57< .0010.096
Licensing/source uncertainty11.4%60.40< .0010.067
Other, please specify8.6%78.20< .0010.076

How to read it

No single chi-square test covers this table. A respondent appears in every row they selected, so the rows are not separate groups of people, and a test of the whole table would count the same person several times. Each answer is tested on its own instead: did the share who selected it differ between the groups? Testing 10 answers gives 10 chances of a false alarm, so the p-values are Holm-adjusted to keep the chance of any false alarm in this table at 0.05.

For “Hard-to-review generated code” the test rejects independence at the 0.05 level (Holm-adjusted p < .001): differences this large between the groups are unlikely if selecting it did not depend on age. Cramér's V of 0.164 is in the small range, which is real but modest: knowing someone's group tells you only a little about their answer.

With 13,648 respondents, even small differences reach statistical significance. Read the percentages and the effect size before the p-value.

Limits

  • The survey has no weights. These tables are unweighted counts of the people who answered, so they describe the respondents, not developers in general.
  • The test treats respondents as a simple random sample. They are people who chose to answer a survey Stack Overflow promoted, not a random sample of developers, so a result holds for these respondents and may not hold beyond them.
  • The adjustment covers the answers of this question by this group only, not the other tables in the library. The headline answer is the strongest of those tested, so its effect size is larger than a typical answer's.
  • A difference between groups is not evidence that group membership causes the answer.

Next steps

Work with this table

Open the test of “Hard-to-review generated code” in the calculator

The link fills in the respondents who selected this answer and those who did not, by group. The whole table does not open there: its rows overlap, and the calculator's test needs rows that are separate people.

Open this question in the full file

The workspace opens all 30,903 respondents in your browser with this question selected. The survey has no weights, so the table there is unweighted too. Change the group, add a filter or export from there.

Open in the workspace →

Run this on your own data — free, no signup

Upload a CSV, Excel, SPSS, Stata, SAS, R, Parquet or JSON file. Everything runs in your browser; your file never leaves your device.

Open the workspace →

The same question by other groups

Other questions that differ by age

Source

Stack Overflow. 2026 Developer Survey, https://survey.stackoverflow.co/2026, licensed under the ODbL 1.0. Tabulated by crosstabs.com. Variables Disruption (AI mistakes that cost time or money) and Age. Stack Overflow licenses the survey data under the Open Database License (ODbL) 1.0 and its contents under the Database Contents License (DbCL) 1.0: credit Stack Overflow, keep the licence notice, and share any database made from the data under the ODbL. These tables and their downloads are offered on those terms.

  1. [1]Stack Overflow: 2026 Developer Survey public results (survey_results_public.csv)survey.stackoverflow.co
  2. [2]Stack Overflow: Open Database License (ODbL) 1.0opendatacommons.org