BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.com//bsides-exeter-2026//speaker//ABDZFW
BEGIN:VTIMEZONE
TZID:GMT
BEGIN:STANDARD
DTSTART:20001029T030000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
TZNAME:GMT
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20000326T020000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:BST
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-bsides-exeter-2026-T3H3JV@pretalx.com
DTSTART;TZID=GMT:20260425T110000
DTEND;TZID=GMT:20260425T112000
DESCRIPTION:AI is rapidly becoming part of the penetration tester’s workf
 low\, generating payloads\, summarising scan results\, and accelerating te
 chnical discovery. But while these tools increase speed\, they also introd
 uce a critical risk: confidence without validation.\n\nAI can suggest vuln
 erabilities that don’t exist\, misinterpret context\, and produce output
  that appears convincing but lacks accuracy. In the hands of an inexperien
 ced tester — or under time pressure — this can lead to false positives
 \, weak findings\, and ultimately poor reporting.\n\nThis talk explores ho
 w AI is actually being used in real-world pentesting\, where it provides g
 enuine value\, and where it can go wrong. Through practical examples\, it 
 highlights common pitfalls and demonstrates how easily unverified AI outpu
 t can make its way into reports.\n\nMore importantly\, it introduces a str
 uctured approach to using AI responsibly\, combining speed with validation
 \, and technical output with real-world context. Attendees will leave with
  a clear framework for integrating AI into their workflow without compromi
 sing credibility or impact.
DTSTAMP:20260501T070549Z
LOCATION:Seminar Room 7
SUMMARY:Autopwn or Auto-Fail? The Truth About AI in Offensive Security - Du
 misani Masimini
URL:https://pretalx.com/bsides-exeter-2026/talk/T3H3JV/
END:VEVENT
END:VCALENDAR