BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.com//state-of-the-map-africa-2025//speaker//V83V
 BD
BEGIN:VTIMEZONE
TZID:EAT
BEGIN:STANDARD
DTSTART:20000101T000000
RRULE:FREQ=YEARLY;BYMONTH=1
TZNAME:EAT
TZOFFSETFROM:+0300
TZOFFSETTO:+0300
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-state-of-the-map-africa-2025-X7YNTX@pretalx.com
DTSTART;TZID=EAT:20260628T140000
DTEND;TZID=EAT:20260628T142000
DESCRIPTION:Mapping has entered a transformative era driven by artificial i
 ntelligence. Traditional geospatial workflows often manual\, time consumin
 g\, and resource-intensive are being fundamentally reimagined through GeoA
 I\, the integration of geospatial data\, science\, and artificial intellig
 ence techniques including machine learning\, deep learning\, and large lan
 guage models (Esri\, n.d.\; Wang et al.\, 2024). This shift is particularl
 y significant for Africa\, where rapid urbanization\, climate resilience n
 eeds\, infrastructure development\, and humanitarian response demand faste
 r\, more accurate\, and scalable mapping solutions. GeoAI automates and au
 gments core mapping tasks such as feature extraction from satellite and ae
 rial imagery\, land cover classification\, change detection\, object detec
 tion\, and predictive spatial modeling (Song et al.\, 2023). By reducing r
 eliance on exhaustive manual digitization\, these technologies help addres
 s persistent challenges in data-scarce environments: limited high-resoluti
 on imagery coverage\, sparse ground-truth data\, and the need for rapid up
 dates in dynamic contexts like informal settlements\, disaster-affected ar
 eas\, and growing cities.\n\nEsri has embedded mature\, production-ready A
 I capabilities throughout the ArcGIS platform\, making advanced GeoAI acce
 ssible to GIS professionals\, government agencies\, NGOs\, and community m
 appers without requiring deep AI expertise. ArcGIS Pro includes the dedica
 ted GeoAI toolbox with tools for training\, fine-tuning\, and deploying mo
 dels on imagery\, point\nclouds\, video\, and tabular data. A growing libr
 ary of over 75 pretrained deep learning models in the ArcGIS Living Atlas 
 enables immediate application for common tasks such as building footprint 
 extraction\, road network detection\, vegetation analysis\, and image segm
 entation — dramatically accelerating base map creation and update cycles
  (Esri\, n.d.).\n\nA major recent advancement is the introduction of AI As
 sistants across the platform. The ArcGIS Pro Assistant (currently in beta)
  allows users to interact with the software using natural language. It can
  generate ArcPy code\, SQL and Arcade expressions\, guide users through co
 mplex workflows\, automate repetitive tasks\, and surface relevant documen
 tation. Complementary assistants in ArcGIS Online and Enterprise support m
 etadata enhancement\, item discovery\, and conversational data exploration
 . These capabilities lower the barrier to advanced analysis and make GIS m
 ore intuitive for a broader range of users\, including those in resource-c
 onstrained settings.\n\nImportantly\, these technologies can complement an
 d strengthen open mapping initiatives such as OpenStreetMap (OSM). AI-powe
 red feature extraction from imagery can pre-process data to suggest candid
 ate features for community mappers to verify and incorporate into OSM\, im
 proving efficiency and coverage. Change detection models can help prioriti
 ze areas requiring updates\, while quality assurance tools can flag potent
 ial inconsistencies. This hybrid approach combining the strengths of autom
 ated GeoAI with human local knowledge and community validation\, offers a 
 powerful pathway for accelerating authoritative and community-driven mappi
 ng across Africa (Iyer et al.\, 2025).\n\nEsri emphasizes Trusted AI princ
 iples transparency\, fairness\, security\, privacy\, and accountability en
 suring that models are responsibly developed and deployed. Pretrained mode
 ls are battle-tested\, and the platform supports fine-tuning on local data
 sets to improve accuracy for African contexts (e.g.\, diverse building typ
 es\, vegetation\, or informal settlement patterns).\n\nLooking ahead\, Esr
 i is advancing geospatial foundation models\, vision-language models GeoVL
 M)\, and location embeddings that further reduce the need for large labele
 d datasets while improving generalization across regions and sensors. Thes
 e developments hold significant promise for the African mapping community 
 (Esri\, n.d.).\n\nThis presentation will demonstrate practical ArcGIS GeoA
 I workflows\, share examples relevant to African use cases (urban mapping\
 , disaster preparedness\, infrastructure monitoring\, and agricultural mon
 itoring)\, and explore opportunities for collaboration between commercial 
 GIS platforms and open mapping communities. By combining the scalability a
 nd analytical power of ArcGIS AI with the ground-truth richness of communi
 ty mapping\, we can collectively accelerate progress toward more complete\
 , current\, and actionable geospatial data for the continent.
DTSTAMP:20260624T135507Z
LOCATION:Audition Room - 2nd Floor -80
SUMMARY:Advancing Mapping Through GeoAI: Esri’s ArcGIS Innovations and Op
 portunities for the African Mapping Community - Ibrahim Mshashi\, Swaleh M
 kangwa:
URL:https://pretalx.com/state-of-the-map-africa-2025/talk/X7YNTX/
END:VEVENT
END:VCALENDAR