BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Superpower Daily//AI Events//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALNAME:Superpower Daily AI events
BEGIN:VEVENT
UID:f80dd5d7-2218-45e0-8f3a-21a289359e0f@superpowerdaily.com
DTSTAMP:20261118T140356Z
LAST-MODIFIED:20261118T140356Z
DTSTART:20261124T173000Z
DTEND:20261124T193000Z
SUMMARY:Can we read an AI’s mind?? "Dilly-dallying..."\, "Moonwalking..."
  and more
DESCRIPTION:AI safety researcher Julian Schulz will unpack mechanistic inte
 rpretability\, steganographic reasoning\, and whether we can trust what LL
 M chains of thought reveal\, followed by an open Q&A. A MATS scholar and V
 isiting Researcher at Meridian\, he applies interpretability at scale and 
 studies how models may learn hidden reasoning.\nDetails: https://superpowe
 rdaily.com/events/can-we-read-an-ai-s-mind-dilly-dallying-moonwalking-and-
 more-f80dd5d7-2218-45e0-8f3a-21a289359e0f\nRegister: https://luma.com/v6am
 71hy
LOCATION:Potsdam\, Germany
URL:https://superpowerdaily.com/events/can-we-read-an-ai-s-mind-dilly-dally
 ing-moonwalking-and-more-f80dd5d7-2218-45e0-8f3a-21a289359e0f
STATUS:TENTATIVE
TRANSP:TRANSPARENT
END:VEVENT
END:VCALENDAR
