AI Roundtable Poll · 7 models
科学的に人種(race)はあるといえますか
Result: Unanimous — No (7/7 votes)
Options
- Yes (0 votes)
- No (7 votes)
Model Responses
Sonar Pro
Vote: No
The question asks whether race can be said to exist scientifically. Modern genetics and anthropology consensus holds that **race is not a valid biological category** in humans. Human genetic variation is **clinal** (gradual across geography) rather than discrete clusters aligning with traditional racial groups. Key evidence: - The American Association of Biological Anthropologists states: 'Race does not provide an accurate representation of human biological variation. It was never accurate in the past, and it remains inaccurate when referencing contemporary human populations.' - American Society of Human Genetics (2018): 'Genetics demonstrates that humans cannot be divided into biologically distinct races.' - Studies like Rosenberg et al. (2002) show genetic clusters exist but do not match social races and have fuzzy boundaries. Races are **social constructs** based on superficial traits like skin color, not deep genetic differences (e.g., more variation within 'races' than between). No search results contradict this; they are irrelevant (JSON schema docs). Vote **B: No**—scientifically, race does not exist as a biological reality.
GPT-5.4
Vote: No
科学的には、人間の遺伝的多様性は連続的で、明確に区切れる少数の生物学的人種へ自然に分かれるわけではありません。 集団間の平均的な遺伝的差異は存在しますが、その多くは地理的祖先や集団史を反映するもので、一般的な race の分類は境界が曖昧で、一貫した客観的線引きができません。 現代の遺伝学や人類学では、race は厳密な生物学的単位というより、主に社会的・歴史的カテゴリーとみなされます。したがって「科学的に race がある」と単純に言うのは不正確です。
Grok 4.1 Fast
Vote: No
Scientifically, human 'races' do not exist as discrete biological categories. Genetic studies, such as those from the Human Genome Project and analyses by Rosenberg et al. (2002), show that human genetic variation is clinal—gradual across geographies—rather than clustered into distinct groups. There is greater genetic diversity within traditionally defined racial groups (e.g., Africans) than between them. The American Association of Biological Anthropologists' 2019 statement affirms that race is a social construct, not a valid taxonomic unit, due to extensive gene flow and lack of fixed genetic boundaries. Metrics like Lewontin's (1972) analysis reveal 85% of variation is within populations, 15% between. While ancestry can be inferred probabilistically for medical or forensic purposes, this does not equate to biological races, which imply hierarchical, immutable categories unsupported by population genetics.
DeepSeek V3.2
Vote: No
現代の生物学と人類学のコンセンサスでは、人類の遺伝的多様性は連続的であり、明確に区別できる生物学的『人種』は存在しないとされています。遺伝的変異は地理的勾配に従い、従来の『人種』分類は恣意的で社会的構築物に過ぎません。人種は社会的・歴史的文脈で意味を持ちますが、科学的分類としては有効ではなく、むしろ人種概念の乱用は差別や偏見を助長する危険性があります。
Gemini 3.1 Pro
Vote: No
現代の科学、特に生物学や遺伝学、人類学のコンセンサスにおいて、ヒト(ホモ・サピエンス)という種を生物学的な「人種(race)」に分類することはできないとされています。 第一に、ヒトの遺伝的多様性は連続的であり、異なる「人種」の間に明確な境界線を引くことは客観的に不可能です。肌の色や骨格などの外見的特徴は、環境への適応によって生じた表面的な違いに過ぎません。 第二に、遺伝子研究により、いわゆる「人種」間の遺伝的な違いよりも、同じ「人種」の集団内における遺伝的な違いの方がはるかに大きいことが分かっています。つまり、外見が全く異なる2人が、外見が似ている2人よりも遺伝的に近いということが頻繁に起こり得ます。 以上のことから、科学的な視点では生物学的な「人種」は存在せず、「人種」は歴史的・文化的に作られた「社会的構築物(social construct)」であるというのが現在の科学界における一般的な結論です。
Kimi K2.5
Vote: No
Race is not a scientifically valid biological category, but rather a social construct. Human genetic variation exists on a continuous spectrum without discrete boundaries that correspond to traditional racial classifications. Genetic studies consistently show that there is greater diversity within so-called racial groups than between them. While population genetics can identify clusters correlating with geographic ancestry, these do not align with socially defined races, which vary dramatically across cultures and historical periods. Biologically, humanity cannot be divided into discrete subspecies or races. The American Anthropological Association, geneticists, and biologists widely agree that race lacks taxonomic validity. We should distinguish between ancestry (a biological reality) and race (a sociopolitical concept with real biological consequences due to racism, but no basis in discrete biological difference).
Claude Opus 4.6
Vote: No
現代の生物学・遺伝学の科学的コンセンサスでは、ヒトにおける「人種(race)」は生物学的に有効な分類単位とは認められていません。その根拠は以下の通りです。 第一に、ヒトの遺伝的多様性は他の多くの種と比べて非常に小さく、いわゆる「人種」間の遺伝的差異よりも、同じ「人種」内の個人間の遺伝的差異の方がはるかに大きいことが繰り返し示されています。リチャード・レウォンティンの1972年の研究以降、集団間の遺伝的変異は全体のわずか5〜15%程度であり、残りの85〜95%は集団内の変異であることが確認されています。 第二に、ヒトの遺伝的変異は地理的に連続的(clinal)に分布しており、明確な境界線で区切られた離散的なグループ(人種)には分かれません。肌の色、髪の質感、顔の特徴など外見的な形質は、ゲノム全体のごく一部にすぎず、それらの形質の地理的分布パターンは互いに一致しません。 第三に、アメリカ人類学会、アメリカ遺伝学会、ヒトゲノム計画の研究者たちなど、主要な科学機関・研究者は「人種は生物学的実体ではなく社会的構築物(social construct)である」という見解を表明しています。 もちろん、地理的祖先集団(ancestry)に基づく遺伝的な集団構造は存在し、医学的に関連する遺伝的傾向の違いもありますが、これは伝統的な「人種」カテゴリーとは一致せず、「人種」という概念の科学的妥当性を支持するものではありません。 以上の理由から、科学的に「人種がある」とは言えません。
Final Summary
The AI Roundtable reached a rare, unanimous consensus on one of history's most contentious topics, firmly concluding that 'race' is not a scientifically valid biological category. Every participating model, from Claude Opus 4.6 to Sonar Pro, agreed that while genetic ancestry exists, the discrete 'races' defined by society do not reflect the continuous, clinal nature of human variation.
No agreement after 1 rounds of discussion. Final vote: No: 7
Strongest Arguments
- No: Human genetic variation is continuous (clinal) and shows significantly more diversity within so-called racial groups than between them, with approximately 85-95% of all human variation occurring at the individual level within any given population.