Best Third Language: French Over Mandarin

A recorded Yes-Brainer council: 3 AI models answering one question, and where they differed. Readable without an API key.

The question

Our 8-year-old speaks English and Spanish and can pick the 3rd language to learn. Converge on ONE for maximum lifetime value.

What each model said

claude-fable-5 (anthropic)

Recommendation: French. I've reconsidered the Mandarin arguments and I'm holding — but let me address the disputed points directly, because they sharpen rather than weaken the case. On the crux — the immersion contingency. All three of us implicitly agree Mandarin only pays off with sustained immersion or heritage support. The user gave us no indication of either. The user asked for ONE answer for maximum lifetime value, and the correct decision rule under uncertainty is to recommend for the base case: a typical bilingual English/Spanish family without Chinese-language infrastructure. In that base case, the realistic outcomes are French fluency by ~age 15 versus Mandarin at intermediate-plat…

gpt-5.5 (openai)

Pick: French — unless your family can realistically provide sustained Mandarin immersion. What changed my mind: my earlier answer over-weighted Mandarin’s ceiling and under-weighted the chance that an 8-year-old actually reaches durable, useful fluency. For “maximum lifetime value,” the better metric is not just payoff; it is: lifetime value = usefulness × probability of real proficiency × likelihood of continued use On that basis, French is the stronger default choice for an English/Spanish-speaking child. Why French wins as the default: 1. Much higher probability of fluency Because your child already has English and Spanish, French will be far easier to acquire than Mandarin. The vocabular…

gemini-3.1-pro-preview (google)

I previously argued for Mandarin based on its unmatched economic ceiling and the cognitive plasticity gained from mastering a structurally distinct language. However, the expected value framework introduced by others—calculating value as payoff multiplied by the probability of reaching fluency—has genuinely changed my mind. Furthermore, the argument regarding machine translation (MT) eroding the transactional value of languages is highly persuasive. I am updating my recommendation to French, establishing a clear conditional framework: Mandarin is only the correct choice if the family already has access to daily, rigorous Mandarin immersion (e.g., a heritage speaker in the home or enrollment…

How it ended

All three participants converged on French as the default third language, framed by expected lifetime value (payoff times probability of fluency), with a shared explicit exception favoring Mandarin only under sustained immersion or heritage conditions. Model A and Model B both reversed their prior Mandarin recommendations to join Model C.

Every answer Yes-Brainer shows is generated by an AI model and can be confidently wrong. Seeing several models agree is not evidence that they are right — the app shows the spread; the judgment stays yours.

Yes-Brainer — ask several AI models one question and see where they disagree. Free and open source; no account, and you bring your own API keys.