引導式測驗 · 中等 等級

人工智慧倫理學家職業道路測驗

測試人工智慧道德工作的各種形式、當前的組織範例、利害關係人分析,以及為什麼標題沒有定義一項標準工作。

問題 1 的 8

What does Google DeepMind’s current careers page say about its Responsibility team?

預覽每個問題8 問題

當您參加測驗時,就會出現答案和解釋。

  1. What does Google DeepMind’s current careers page say about its Responsibility team?
    • It replaces product and research teams when making launch decisions
    • It spans policy, research, and engineering and addresses societal impacts
    • It is limited to reviewing legal contracts
    • It only writes model training code
  2. Which activity appears in OpenAI’s current Product Policy Research and Advisory Partnerships posting?
    • Set every model’s technical architecture independently
    • Handle only customer support escalations
    • Train production models without engaging external experts
    • Build advisory relationships and translate external research into guidance for product and policy teams
  3. How should readers characterize the job title “AI ethicist”?
    • It can refer to related work placed in different research, policy, product, or governance roles
    • It has one universally standardized credential and career ladder
    • It always means a legal-compliance officer
    • It always means an ML engineer who tests model accuracy
  4. What information belongs in a responsible-AI impact assessment?
    • Only the legal department’s final approval
    • Only the number of users who tried a prototype
    • Affected groups, plausible outcomes, evidence gaps, decision owners, and possible mitigations
    • Only the model’s benchmark score
  5. How does ethical analysis relate to legal compliance and technical safety testing?
    • It proves a model is technically reliable
    • It guarantees a use case has no harmful effects
    • It can complement those practices but does not replace them
    • It makes legal review unnecessary
  6. Which candidate evidence best fits an interdisciplinary AI ethics role?
    • Clear analysis and writing that connects stakeholder evidence to technical or policy decisions
    • A model demo with no explanation of affected users
    • A claim that a checklist settles every value judgment
    • A salary survey without examples of research or collaboration
  7. What should an applicant ask about a specific ethics-related role?
    • Whether all ethics questions can be scored by one metric
    • Whether the employer uses one universal title
    • Whether a degree guarantees decision authority
    • How recommendations reach design or deployment decisions and who acts on them
  8. Why should role descriptions be read individually?
    • Employers cannot combine ethics and engineering work
    • The scope may center on social research, policy, technical evaluations, or stakeholder engagement depending on the team
    • Every ethics role has the same tasks but different names
    • The title alone guarantees authority over all launches