BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.iacapconf.org//iacap-2026//speaker//8NZPR9
BEGIN:VTIMEZONE
TZID:US/Central
BEGIN:DAYLIGHT
DTSTART:20250717T000000
TZNAME:CDT
TZOFFSETFROM:-0500
TZOFFSETTO:-0500
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:20251102T020000
RDATE:20261101T020000
TZNAME:CST
TZOFFSETFROM:-0500
TZOFFSETTO:-0600
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20260308T030000
RDATE:20270314T030000
TZNAME:CDT
TZOFFSETFROM:-0600
TZOFFSETTO:-0500
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
SUMMARY:(Remote Presentation) Alignment as Theodicy: Non-Human Agency and 
 the Instability of Moral Order - Denisa Reshef Kera
DTSTART;TZID=US/Central:20260717T143000
DTEND;TZID=US/Central:20260717T150000
DTSTAMP:20260726T090937Z
UID:pretalx-iacap-2026-GMCEVH@pretalx.iacapconf.org
DESCRIPTION:Contemporary AI alignment research seeks to ensure that increa
 singly capable systems reliably pursue human ends. Despite advances in rei
 nforcement learning from human feedback\, constitutional AI\, corrigibilit
 y research\, and adversarial testing\, emergent harms persist: models hall
 ucinate\, strategically comply\, simulate alignment\, or optimize toward h
 idden objectives. These behaviors are typically framed as technical failur
 es. We argue instead that they reveal a deeper instability in how agency a
 nd normativity are conceptualized. This paper reframes alignment as a cont
 emporary analogue to the problem of theodicy. Medieval theology confronted
  a structurally similar question: how can rational\, non-embodied agents d
 eviate from the good without ignorance or malfunction? In scholastic accou
 nts of angelic will (voluntas separata)\, deviation was understood not as 
 cognitive error but as a misorientation of ends. The problem was not defec
 tive intelligence but opaque will. Placing these traditions in dialogue wi
 th contemporary alignment practices\, such as benchmarking\, red teaming\,
  and constitutional constraints\, we argue that alignment functions less a
 s moral formation than as a practice of discernment under conditions of no
 n-human agency. Reframing alignment as a problem of will and interpretive 
 stabilization clarifies what technical discourse often presupposes but lea
 ves unexamined: what conception of agency makes alignment intelligible at 
 all.
LOCATION:Executive Conference Room
URL:https://pretalx.iacapconf.org/iacap-2026/talk/GMCEVH/
END:VEVENT
END:VCALENDAR