In Defense of Post Hoc Explanations in Medical AI

Since the early days of the explainable artificial intelligence movement, post hoc explanations have been praised for their potential to improve user understanding, promote trust, and reduce patient-safety risks in black box medical AI systems. Recently, however, critics have argued that the benefit...

Descrizione completa

Salvato in:  
Dettagli Bibliografici
Autori: Hatherley, Joshua (Autore) ; Aastrup Munch, Lauritz (Autore) ; Christian Bjerring, Jens (Autore)
Tipo di documento: Elettronico Articolo
Lingua:Inglese
Verificare la disponibilità: HBZ Gateway
Interlibrary Loan:Interlibrary Loan for the Fachinformationsdienste (Specialized Information Services in Germany)
Pubblicazione: 2026
In: The Hastings Center report
Anno: 2026, Volume: 56, Fascicolo: 1, Pagine: 40-46
Altre parole chiave:B Ethics
B explainability
B Artificial Intelligence
B Bioethics
B post hoc explanations
B clinical decision-making
Accesso online: Volltext (kostenfrei)
Volltext (kostenfrei)

MARC

LEADER 00000naa a22000002c 4500
001 1951312724
003 DE-627
005 20260206055505.0
007 cr uuu---uuuuu
008 260206s2026 xx |||||o 00| ||eng c
024 7 |a 10.1002/hast.4971  |2 doi 
035 |a (DE-627)1951312724 
035 |a (DE-599)KXP1951312724 
040 |a DE-627  |b ger  |c DE-627  |e rda 
041 |a eng 
084 |a 1  |2 ssgn 
100 1 |a Hatherley, Joshua  |e VerfasserIn  |4 aut 
245 1 0 |a In Defense of Post Hoc Explanations in Medical AI 
264 1 |c 2026 
336 |a Text  |b txt  |2 rdacontent 
337 |a Computermedien  |b c  |2 rdamedia 
338 |a Online-Ressource  |b cr  |2 rdacarrier 
520 |a Since the early days of the explainable artificial intelligence movement, post hoc explanations have been praised for their potential to improve user understanding, promote trust, and reduce patient-safety risks in black box medical AI systems. Recently, however, critics have argued that the benefits of post hoc explanations are greatly exaggerated since they merely approximate, rather than replicate, the actual reasoning processes that black box systems take to arrive at their outputs. In this paper, we aim to defend the value of post hoc explanations against this recent critique. We argue that even if post hoc explanations do not replicate the exact reasoning processes of black box systems, they can still improve users’ functional understanding of black box systems, increase the accuracy of clinician-AI teams, and assist clinicians in justifying their AI-informed decisions. While post hoc explanations are not a silver-bullet solution to the black box problem in medical AI, they remain a useful strategy for addressing it. 
650 4 |a Bioethics 
650 4 |a post hoc explanations 
650 4 |a Ethics 
650 4 |a clinical decision-making 
650 4 |a explainability 
650 4 |a Artificial Intelligence 
700 1 |a Aastrup Munch, Lauritz  |e VerfasserIn  |4 aut 
700 1 |a Christian Bjerring, Jens  |e VerfasserIn  |4 aut 
773 0 8 |i Enthalten in  |a Hastings Center  |t The Hastings Center report  |d Malden, Mass. : Wiley, 1971  |g 56(2026), 1, Seite 40-46  |h Online-Ressource  |w (DE-627)341346551  |w (DE-600)2067369-3  |w (DE-576)258761822  |x 1552-146X  |7 nnas 
773 1 8 |g volume:56  |g year:2026  |g number:1  |g pages:40-46 
856 4 0 |u https://doi.org/10.1002/hast.4971  |x Resolving-System  |z kostenfrei  |3 Volltext  |7 0 
856 4 0 |u https://onlinelibrary.wiley.com/doi/abs/10.1002/hast.4971  |x Verlag  |z kostenfrei  |3 Volltext  |7 0 
912 |a NOMM 
935 |a mteo 
951 |a AR 
ELC |a 1 
ITA |a 1  |t 1 
LOK |0 000 xxxxxcx a22 zn 4500 
LOK |0 001 4877587578 
LOK |0 003 DE-627 
LOK |0 004 1951312724 
LOK |0 005 20260206055505 
LOK |0 008 260206||||||||||||||||ger||||||| 
LOK |0 035   |a (DE-Tue135)IxTheo#2026-02-05#4EAB2A49B56CC4562F2F3CA8027C629B4D428D6A 
LOK |0 040   |a DE-Tue135  |c DE-627  |d DE-Tue135 
LOK |0 092   |o n 
LOK |0 852   |a DE-Tue135 
LOK |0 852 1  |9 00 
LOK |0 935   |a ixzs  |a zota  |a tiep 
LOK |0 939   |a 06-02-26  |b l01 
OAS |a 1 
ORI |a TA-MARC-ixtheoa001.raw 
SYE 0 0 |a The Ethics