Laila Shereen Sakr
Wednesday, March 25, 2026

Department of AI and Society
University at Buffalo

A Contrapuntal Approach to Global Data

Genealogy

M.A. in Foreign Service (Arab Studies)

Transdisciplinary Background

Researcher, Builder, Artist

• AI + Digital Humanities Infrastructure

• R-Shief → Contrapuntal

• VJ Um Amel (Arabic for VJ "Mother of Hope")

"Rosetta Stones Resurrected," a 3D hologram (2025)

R-Shief Media System Homepage Interface (2016)

"Intro to Arab, Mother, Cyborg Digital Performance" (2010)

Roadmap

PROBLEM

• Global data is uneven, multilingual, power-structured

• AI systems
  flatten context

• Result: loss of meaning, misrepresentation

METHOD

• Glitch → reading systems against the grain
• Contrapuntal → holding multiple histories in relation
• Culturally aware ontological approach

SYSTEM

• R-Shief (archive → analysis)
• Contrapuntal AI (design → intervention)

CONTRIBUTION

• New dataset design principles
• Humanities inside ML pipelines
• AI as knowledge infrastructure

 R-Shief → Arabic Glitch → Contrapuntal

(2009+)       (2023)              (2025+)

Research Question

For example, sentiment analysis fails to recognize sarcastic humor (2012).

How can AI systems represent global data without flattening the historical, linguistic, and political relations that give it meaning?

Challenge
Global data is uneven, multilingual, power-structured and AI flattens context.

• 70+ languages

• billions of posts

• global archive of social media

R-Shief (2009+)

R-Shief Cultural Analytics

#WomensMarch, 2016

#OccupyWallStreet, 2011

Hong Kong #Umbrella, 2012

#BlackLivesMatter, 2014

Data extraction was not enough

• Pattern ≠ meaning

• Scale ≠ interpretation

• Multilingual data ≠ contextual understanding

• Archive required new epistemology

What R-Shief Taught Me

Why the Middle East context matters now

• Language underrepresentation

• Regional AI investment

• Data centers, models, infrastructure

• Geopolitics of AI

Middle East AI and

Global Media Power

International Journal of Middle East Studies, Tina Guirguis, 2025-01-20, “Arabic Glitch … deftly remixes and traces the history of the Arabization of the Internet.”

International Journal of Communication, Nathaniel Greenberg, 2026-02,   “Arabic Glitch...emerges as a shining academic achievement in a decades-long scramble by academics, journalists, and artists to capture some of that creativity and to tell the story of the Arab uprisings as one of rapid technological change, political reactionism, and creative resistance.”    

Journal of Cyber Policy, Jessica McClearn, 2024-07-18, “This text will serve as a guidepost for future exploratory work across disciplines.”

Neural, Neural Editors, 2024-07-19, “The glitch practise she proposes thus becomes a methodology of awareness and liberation …”    

Sample Reviews from Flagship Journals

Arabic Glitch Interventions

1. Glitch as method

2. Data body (no virtual/real divide)

3. Procedural literacy

From Media Systems → AI Systems

• Data Centered Collective 
  Co-Founder, along with Miha Vipotnik (Slovenia/Lebanon), Temitayo Ogunbiyi (Nigeria), and Lisa Parks (US)
   
• Network of Arab Women in AI
  Co-Founder, along with colleagues in Egypt, Lebanon, Germany, Canada, and Qatar
   
• Autonomous Futures
  Co-Founder, along with colleagues at Michigan Central, Ohio State University, META, and Arizona State University

Collaborative Initiatives (2021+)

Contrapuntal

• multiple histories

• in relation

• without resolution

Capitol Mosaic, 30-ft wall of Jan 6 images from Parler and Twitter, 
VJ Um Amel (Qualcomm Gallery, 2021).

From Archive to AI System:
Contrapuntal AI

• Human–AI collaborative platform

• Contextual metadata

• Citational lineage

• Participatory annotation

• Small language models

• Interpretability + bias evaluation
   

Methodology

1. Data Layer

• contextual metadata

• multilingual structure

• historical annotation

2. Model Layer

• small language models

• comparative training (baseline vs enriched datasets)

3. Evaluation Layer

• interpretability

• bias mitigation

• semantic fidelity across languages

Participatory Design

Public & Creative Scholarship

Selection

“When Data Bodies Become Targets: New Risks in a Networked World.” Tech Policy Press, 4 Nov. 2024,

“Rosetta Stones Resurrected.” Art Must Be Artificial: Perspectives of AI in the Visual Arts, curated by Jérôme Neutres, Diriyah Art Futures, Riyadh, Saudi Arabia, 26 Nov. 2024–15 Feb. 2025. 

Live audio-visual performance using social media data via R-Shief. Presented at ISEA International Symposium on Electronic Art, Vancouver, BC, 2016.

"Network Visualization on 2 million X posts on #MahsaAmini" (November 2022)

Publications

Selection

Current Projects

Contrapuntal Futures: From Social Media to
AI-Generated Information,

MIT Press

(forthcoming).

Selection

Courses

Selection

 Why UB / Why AI + Society

Why it matters

• AI as global media infrastructure

• Cross-disciplinary collaboration

• Research + pedagogy + public engagement

• Plural worlds need plural methods

Acknowledgements

Ripley Baker
Pratyush 'Rumi' Bhattacharyya
Henry Coburn
Lexxus Edison
Negar Farajiani
Tina Guirguis
Intae Hwang

Susana Ruiz (Co-PI)
Cathy Thomas (Co-PI)
Alenda Chang (Co-Director)
Data Centered Collective:
Miha Vipotnik (Slovenia/Lebanon), Temitayo Ogunbiyi (Nigeria), and Lisa Parks (US)

 

Access to Knowledge for
Development (AUC, Egypt):
Nagla Rizk
Autonomous Futures
Network of Arab Women in AI
Creative Critical Undercommons
Arab Techies

Cass Mayeda
Sihwa Park
Chelsea Kai Roesch

Shashank Shivashankar