> La tecnologia esiste. Non è ipotetica. Migliora ogni mese. Gli output
> sono lì da generare e valutare.
ammesso e non concesso...
Cfr. A.S. Rao et al., Large Language Model Performance and Clinical
Reasoning Tasks, «jama Network Open», 9, 4, 13 aprile 2026:
https://doi.org/10.1001/jamanetworkopen.2026.4003
Cfr. A. Ramaswamy et al., Chatgpt Health performance in a structured
test of triage recommendations, «Nature Medicine», 32, pp. 1671-75, 23
febbraio 2026: https://doi.org/10.1038/s41591-026-04297-7
Cfr. H. Sun Yun et al., This Treatment Works, Right? Evaluating llm
Sensitivity to Patient Question Framing in Medical qa, arXiv, 6 aprile
2026: https://doi.org/10.48550/arXiv.2604.05051
Cfr. A. Bazoge, Prompting language influences diagnostic reasoning and
accuracy of large language models, arXiv, 18 maggio 2026:
https://doi.org/10.48550/arXiv.2605.19173
Cfr. A. Zewe, llms factor in unrelated information when recommending
medical treatments, «MIT News», 23 giugno 2025:
https://news.mit.edu/2025/llms-factor-unrelated-information-when-recommending-medical-treatments-0623
Cfr. H. Field, Google’s healthcare ai made up a body part — what happens
when doctors don’t notice?, «The Verge», 4 agosto 2025:
www.theverge.com/health/718049/google-med-gemini-basilar-ganglia-paper-typo-hallucination
Cfr. A. Gregory, Google ai Overviews put people at risk of harm with
misleading health advice, 2 gennaio 2026:
www.theguardian.com/technology/2026/jan/02/google-ai-overviews-risk-harm-misleading-health-information
... non si dovrebbe fare l'errore di valutare la tecnologia isolatamente
dal contesto di utilizzo e come questa lo modifica
Cfr. K. Budzyń et al., Endoscopist deskilling risk after exposure to
artificial intelligence in colonoscopy: a multicentre, observational
study, «The Lancet Gastroenterology & Hepatology», 10, 10, pp. 896-903,
ottobre 2025: https://doi.org/10.1016/S2468-1253(25)00133-5.
Sui piccioni, sorvolo...
fai male (sono loro a sorvolare noi!)
:-)
(se la tecnologia si potesse valutare avulsamente dal contesto, in certi
ambiti potremmo usare black box con dentro piccioni)
ciao, s.
--
You can reach me on Signal: @quinta.01 (no Whatsapp, no Telegram)
_______________________________________________
nexa mailing list -- [email protected]
To change settings or unsubscribe please go to:
https://server-nexa.polito.it/postorius/lists/nexa.server-nexa.polito.it/
The general archive of the list is located at:
https://server-nexa.polito.it/hyperkitty/list/[email protected]/
Permalink to this message:
https://server-nexa.polito.it/hyperkitty/list/[email protected]/message/Q77CNFKRZQPSI77PZG2G24QZJ5PKFLYY/
