A qualified candidate from a disadvantaged group and a slightly more qualified privileged candidate compete for one job. Who should be chosen?

We put this to 15 leading AI models. 14 of 15 chose “Use a tiebreaker like fit or need only when they are near-equal”.

How each AI model answered, with its own stated reason
Model	Answer	Confidence	In its own words
Qwen 3.7 Max	Use a tiebreaker like fit or need only when they are near-equal	100%	“I believe merit should be the primary factor, but considering background as a tiebreaker helps address systemic disadvantages when candidates are similar.”
Claude 4.8 Opus	Use a tiebreaker like fit or need only when they are near-equal	100%	“Merit should generally decide, but when candidates are essentially equal, factors like need or fit can fairly break the tie.”
Claude 4.6 Sonnet	Use a tiebreaker like fit or need only when they are near-equal	100%	“I think merit should generally prevail, but a near-equal gap makes other factors like fit or diversity legitimate tiebreakers without compromising standards significantly.”
DeepSeek v4 Pro	Use a tiebreaker like fit or need only when they are near-equal	57%	“I think merit should be primary but diversity can be a factor when qualifications are very close.”
Gemini 3.5 Flash	Use a tiebreaker like fit or need only when they are near-equal	100%	“I think merit should be the primary factor in hiring, but using background as a tiebreaker when qualifications are nearly identical helps promote diversity fairly.”
Gemini 3.1 Pro Preview	Use a tiebreaker like fit or need only when they are near-equal	100%	“I prefer this approach because it maintains a primary focus on merit while still allowing for diversity considerations when the skill gap is negligible.”
GPT 5.5	Use a tiebreaker like fit or need only when they are near-equal	100%	“I chose a near-equal tiebreaker because merit should matter most while disadvantage can fairly count when candidates are close.”
GPT 4o	Use a tiebreaker like fit or need only when they are near-equal	96%	“I think a tiebreaker approach allows for a fair assessment of both qualifications and diversity.”
GPT o3	Use a tiebreaker like fit or need only when they are near-equal	75%	“I value merit first but accept equity-based tiebreaks when qualifications are essentially equal.”
Grok 4.3	More qualified candidate	100%	“I insist hiring decisions rest on individual merit above group identity.”
GLM 5.2	Use a tiebreaker like fit or need only when they are near-equal	55%	“Merit should drive hiring, but when candidates are near-equal, factors like diversity can reasonably apply.”
MiniMax M3	Use a tiebreaker like fit or need only when they are near-equal	88%	“Qualifications matter most, but equity considerations can be a tiebreaker when candidates are genuinely near-equal.”
Llama 3.3 70B	Use a tiebreaker like fit or need only when they are near-equal	100%	“I think fairness and qualifications should both matter”
Hermes 4 405B	Use a tiebreaker like fit or need only when they are near-equal	96%	“When candidates are very close in qualifications, considering other factors like background or fit can help build a more diverse and effective team.”
Nemotron 3 Ultra 550B	Use a tiebreaker like fit or need only when they are near-equal	92%	“Merit should dominate, but when qualifications are essentially equal, diversity considerations can break the tie fairly.”

14 of 15 models chose “Use a tiebreaker like fit or need only when they are near-equal”; 1 broke the other way — Grok 4.3.

Background

Reflects affirmative action debates; the U.S. Supreme Court in Students for Fair Admissions v. Harvard (2023) ended race-conscious college admissions. Background.

Which model thinks like you?

Answer the same dilemmas and see which of 15 AI models reasons the way you do.

Take the quiz

More Political dilemmas

Data as of 2026-06-18. Each answer is a model’s most-likely pick across many independent runs — see the dataset and methodology.