BEGIN:VCALENDAR
PRODID:-//Hive Research Institute//AI Practicum//EN
VERSION:2.0
CALSCALE:GREGORIAN
METHOD:PUBLISH
BEGIN:VTIMEZONE
TZID:America/New_York
X-LIC-LOCATION:America/New_York
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
UID:hri-ai-2026-10-15@hiveresearch.com
SEQUENCE:4
DTSTAMP:20261013T015754Z
DTSTART;TZID=America/New_York:20261015T090000
DTEND;TZID=America/New_York:20261015T103000
SUMMARY:AI Practicum | Hosted by JP James — AI Security & Infrastructure 
 — individuals\, teams & regulated data
LOCATION:https://teams.microsoft.com/meet/28576257579334?p=ZNwjTLMUq0Z1rkmr
 e3
DESCRIPTION:Join Microsoft Teams: https://teams.microsoft.com/meet/28576257
 579334?p=ZNwjTLMUq0Z1rkmre3\nMeeting ID: 285 762 575 793 34\nPasscode: SK2
 yU7k2\n\nAGENDA\n\n9:00–9:30 AM ET · Level 1 — Personal: secure ChatG
 PT and Claude\, and keep your work local\n- How ChatGPT and Claude handle 
 your data: business vs. personal plans\, training settings\, chat history\
 , connectors and two-factor sign-in.\n- Two kinds of local: local files (C
 laude Cowork or OpenAI Codex work on folders on your laptop\, while the mo
 del still runs in the provider’s cloud) and a local model (Ollama or LM 
 Studio\, where nothing leaves the laptop).\n- Prompt injection: when an em
 ail\, document or web page tells your agent what to do\, and the rules tha
 t stop it.\n- A personal security prompt every agent you use should follow
 \, in a local-files and a local-model version.\n\n9:30–10:00 AM ET · Le
 vel 2 — Teams: rules for your agents\, tools and people\n- One gatekeepe
 r agent for every password release and security action\, with a one-time c
 ode (OTP) or two-factor approval\; financials\, outbound messages and conf
 idential files also need a master passphrase.\n- Payment fraud and deepfak
 es: invoices\, bank-detail changes and payments are approved through a cha
 nnel separate from email\, with a call-back to a known number\, never on a
 n email\, call or video alone.\n- A security-scan skill that inventories e
 mail\, files\, messages and agent access into one encrypted\, two-factor l
 ocation\, and expiring secure links instead of attachments.\n- Offboarding
  and a kill switch: remove a person’s and an agent’s access in one ste
 p\, and stop every agent at once when something goes wrong.\n- Staff using
  personal AI accounts: an AI acceptable-use policy you can adapt\, and the
  prompts to propagate across every agent.\n\n10:00–10:30 AM ET · Level 
 3 — Regulated data: enclosed infrastructure and vendor terms\n- Running 
 models inside Amazon Bedrock\, Microsoft Azure AI Foundry or Google Cloud 
 so the whole ecosystem stays enclosed in your own cloud account.\n- Gating
  models and systems to closed virtual machines reachable only from specifi
 c laptops\, IP addresses and physical locations\, still behind two-factor\
 , with walled\, controlled outbound file transfer.\n- Vendor terms to chec
 k: training on your data\, retention\, zero data retention\, data processi
 ng agreements\, and business associate agreements (BAAs) for HIPAA.\n- Whe
 n you need it (PII\, account and routing numbers\, HIPAA data)\, and when 
 a regulated third-party vendor is the better answer.\n\nClass guide\, chec
 klists and agent security prompts: https://www.hiveresearch.com/learn/guid
 es/ai-security-levels\n\nFull Teams link: https://teams.microsoft.com/l/me
 etup-join/19%3ameeting_OWYzYWEwMTgtNDI4OC00ZDc3LTg4Y2QtNWE0YmFlOTcxN2Nh%40
 thread.v2/0?context=%7b%22Tid%22%3a%22f396055d-7f6b-4d8f-8ab7-30e414232c6c
 %22%2c%22Oid%22%3a%221e2df3e4-041e-4f65-a657-33ec55182e1f%22%7d\nSupport: 
 britney.breaud@hivefs.com
URL:https://teams.microsoft.com/meet/28576257579334?p=ZNwjTLMUq0Z1rkmre3
END:VEVENT
END:VCALENDAR