BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.hackglasgow.live//hack-glasgow-2026//talk//LJDXM
 3
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-hack-glasgow-2026-LJDXM3@pretalx.hackglasgow.live
DTSTART;TZID=GMT:20260815T163000
DTEND;TZID=GMT:20260815T165500
DESCRIPTION:The Commonality Problem in AI-Assisted Offensive Code Develeope
 mnt. \n\nSince the public debut of ChatGPT\, the security community has co
 ntinued to focus on the wrong question. The issue is not whether AI can ge
 nerate malware that capability is already established. More importantly\, 
 it is not the most significant development.\n\nThis talk focus three evide
 nce-based ways genAI is reshaping the offensive security landscape\, along
  with a fourth emerging risk that is closer than many defenders assume.\n\
 nFirst is code convergence. When different threat actors rely on the same 
 genAI systems\, the resulting malicious code begins to show structural sim
 ilarities. This is not due to coordination\, but to shared training data a
 nd model behavior. As a result\, detection systems designed tend end to ca
 tch only lower-skill actors\, while more advanced operators evade detectio
 n. \n\nSecond is novel technique synthesis. Similar to how AI in drug disc
 overy evolved from searching known compounds to generating entirely new on
 es\, genAI is likely to produce offensive techniques that do not exist in 
 current datasets. Evidence from various research initiatives shows the int
 egrations with LLMs\, and academic research into automated exploit generat
 ion supports this shift from replication to creation.\n\nThird is the two-
 world problem. genAI does not impact all threat actors equally. Disclosure
 s from Microsoft and OpenAI identified multiple state-linked groups and ac
 tors using LLMs o support offensive activity. This talk analyses what each
  tier gains\, what each tier does not\, and why a single defensive respons
 e to “the AI threat” is already insufficient.
DTSTAMP:20260611T152322Z
LOCATION:Stage 2
SUMMARY:The Model Knows What Works. They All Asked the Same Thing. - Jinto 
 Antony
URL:https://pretalx.hackglasgow.live/hack-glasgow-2026/talk/LJDXM3/
END:VEVENT
END:VCALENDAR