BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//cfp.ducttape.events//ethprague-2026//talk//9RFDE8
BEGIN:VTIMEZONE
TZID:CET
BEGIN:STANDARD
DTSTART:20001029T040000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
TZNAME:CET
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20000326T030000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:CEST
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-ethprague-2026-9RFDE8@cfp.ducttape.events
DTSTART;TZID=CET:20260508T142000
DTEND;TZID=CET:20260508T144500
DESCRIPTION:We built Wake Arena — multi-agent AI with graph-driven reason
 ing — and benchmarked it against 94 real vulnerabilities from audit comp
 etitions alongside plain GPT\, plain Opus\, and AI security services. I'll
  share the numbers\, why graph context wins over LLM wrappers\, and where 
 AI approaches still fall short in real audits.
DTSTAMP:20261017T235721Z
LOCATION:Grégr Stage
SUMMARY:Benchmarking AI for Smart Contract Security: What Actually Works - 
 Josef Gattermayer
URL:https://cfp.ducttape.events/ethprague-2026/talk/9RFDE8/
END:VEVENT
END:VCALENDAR