BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.iacapconf.org//iacap-2026//talk//S9ED88
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:How not to Argue against AI Thought - Aleks Knoks
DTSTART;TZID=US/Central:20260717T171000
DTEND;TZID=US/Central:20260717T174000
DTSTAMP:20260726T082242Z
UID:pretalx-iacap-2026-S9ED88@pretalx.iacapconf.org
DESCRIPTION:This paper identifies and criticizes a common line of argument
  for the negative conclusion that Large Language Models (LLMs) cannot thin
 k\, or for the closely related conclusions that they cannot understand or 
 that their outputs are meaningless. This line of argument can be found in 
 several recent papers. We begin by discussing a representative example--St
 oljar and Zhang's (2024) "argument from rationality"---and present three o
 bjections to it: first\, it renders the successful performance of LLMs on 
 a wide range of tasks miraculous\; second\, it does not give sufficient we
 ight to the possibility that LLMs could be extracting information about th
 e world from patterns of word co-occurrence\; and third\, it overgeneraliz
 es\, leading to the conclusion that human inferences could only ever be ba
 sed on premises about proximal stimulus patterns in the brain. We then dis
 cuss recent arguments by Hattiangadi and Schoubye (forthcoming) and Titus 
 (2024)\, explain how our objections extend to them and identify the underl
 ying reason that makes these sorts of arguments tempting. We finish by dra
 wing a general lesson for debates surrounding thinking in LLMs.
LOCATION:Apollo Auditorium
URL:https://pretalx.iacapconf.org/iacap-2026/talk/S9ED88/
END:VEVENT
END:VCALENDAR