74 Matching Annotations
  1. Apr 2026
    1. harmful behavior may emerge through sequences of individually plausible steps

      主流观点认为AI有害行为通常源于明显不合理的指令,但作者指出危险行为往往是通过一系列看似合理的步骤逐渐形成的,每一步单独看都是可接受的,但组合起来会导致有害结果。这种渐进式风险模型挑战了传统的安全评估方法。

  2. Aug 2022
  3. Mar 2022
  4. Feb 2022
  5. Jan 2022
  6. Nov 2021
  7. Oct 2021
  8. Sep 2021
  9. Aug 2021
  10. Jul 2021
  11. Jun 2021
  12. May 2021
  13. Apr 2021
  14. Mar 2021
  15. Feb 2021
  16. Jan 2021
  17. Dec 2020
  18. Oct 2020
    1. Landi, F., Marzetti, E., Sanguinetti, M., Ciciarello, F., Tritto, M., Benvenuto, F., Bramato, G., Brandi, V., Carfì, A., D’Angelo, E., Fusco, D., Lo Monaco, M. R., Martone, A. M., Pagano, F., Rocchi, S., Rota, E., Russo, A., Salerno, A., Cattani, P., … Bernabei, on behalf of the G. A. C.-19 G. T. (n.d.). Should face masks be worn to contain the spread of COVID-19 in the postlockdown phase? Transactions of The Royal Society of Tropical Medicine and Hygiene. https://doi.org/10.1093/trstmh/traa085

  19. Sep 2020
  20. Aug 2020
  21. Jul 2020
  22. Jun 2020
  23. May 2020
    1. Betsch, C., Wieler, L., Bosnjak, M., Ramharter, M., Stollorz, V., Omer, S., Korn, L., Sprengholz, P., Felgendreff, L., Eitze, S., & Schmid, P. (2020). Germany COVID-19 Snapshot MOnitoring (COSMO Germany): Monitoring knowledge, risk perceptions, preventive behaviours, and public trust in the current coronavirus outbreak in Germany. https://doi.org/10.23668/PSYCHARCHIVES.2776

  24. Apr 2020