-
Notifications
You must be signed in to change notification settings - Fork 0
5.7 The Datafied Present and Future: A Guide for Academic Practice
Content adapted from: Atenas, J. (2021). The Datafied Present and Future. In Understanding Data: Praxis and politics. HDI - Data, Praxis and Politics. https://doi.org/10.5281/zenodo.4698609
We are living in a moment where data is not simply descriptive—it is constitutive of reality. Data infrastructures increasingly shape how decisions are made, how people are represented, and how futures are predicted. For academics, this requires a shift from teaching data as a technical artefact towards understanding it as a site of power, ethics, and contestation.
This guide offers a framework for embedding critical, ethical, and socially engaged approaches to data within academic practice.
In the datafied present, nearly all aspects of life are translated into data. These processes are deeply embedded within social, economic, and political systems, influencing how individuals are categorised, governed, and understood. A key concern is the interconnected nature of data infrastructures, which increasingly enable predictive logics:
The interwovenness of data infrastructures facilitate attempts to predict socioeconomic behaviours, by promoting the collection of socioeconomic data (race, gender, neighbourhood) aiming to predict certain behaviours depending on people’s background… from educational success to insurance pricing and predictive policing.
This raises critical questions about:
- How data is used to categorise individuals
- How prediction shapes opportunity and risk
- How inequalities are reproduced through data systems
Academic implications
- Teach students to question predictive systems
- Highlight how data classifications impact lived realities
- Explore how data infrastructures reinforce structural inequalities
flowchart TD
A["Human Activity<br/>Social, economic, educational"] --> B[Data Generation]
B --> C[Data Collection & Storage]
C --> D["Data Infrastructures<br/>Platforms, institutions"]
D --> E[Analysis & AI Systems]
E --> F[Inference & Prediction]
F --> G1[Behavioural Influence]
F --> G2[Decision-Making Systems]
F --> G3[Classification & Profiling]
G1 --> H[Individual Behaviour Changes]
G2 --> H
G3 --> H
H --> I["Feedback Loop<br/>New Data Generated"]
I --> B
%% Inequality & Risk layer
E --> J[Bias & Opacity]
J --> K[Social Inequality]
K --> L1[Racism]
K --> L2[Gender Bias]
K --> L3[Socioeconomic Inequality]
%% Governance
E --> M[Ethics & Regulation]
M --> N[Data Protection & AI Governance]
%% Styling (pastel + black text)
classDef pastel fill:#f5e6ff,stroke:#555,color:#000;
classDef pastel2 fill:#e6f7ff,stroke:#555,color:#000;
classDef pastel3 fill:#e8ffe6,stroke:#555,color:#000;
class A,B,C,D pastel;
class E,F,G1,G2,G3 pastel2;
class H,I pastel3;
class J,K,L1,L2,L3 pastel;
class M,N pastel2;
Data ethics provides a foundation for understanding the moral dimensions of data practices. It frames decisions about what should be collected, how it should be used, and whose interests it serves. Ethics must be understood not as a compliance exercise, but as a critical and reflexive practice embedded throughout the data lifecycle
- Ethical reflection as part of all research
- Responsibility towards individuals and communities
- Awareness of harm, bias, and unintended consequences
Ethical engagement requires questioning:
- Who benefits from data use
- Who is excluded or harmed
- How power shapes data narratives
AI, algorithms, and machine learning are transforming decision-making at scale, reshaping institutions, governance, and everyday life. However, these technologies raise fundamental concerns:
AI presents risks and challenges to legal systems, particularly due to its opacity and bias. Automated decision-making systems often lack accountability, community engagement, and auditing, creating power imbalances and limiting opportunities for participation.
A central issue is opacity:
Individuals affected by algorithmic decisions often cannot understand or challenge them. Decisions appear objective but may embed hidden assumptions
Addressing this requires:
Reducing bias Increasing transparency Developing appropriate legal frameworks
To grasp how AI affects everyday life, we must understand how data flows through systems. The Human–Data Interaction model illustrates that:
- Individuals generate data
- Systems analyse data to produce inferences
- These inferences shape behaviour
- Behaviour generates new data
This creates a continuous feedback loop, where inferences derived from personal data are fed back into systems, influencing future behaviour and becoming subjects of further analysis
In the datafied present, teaching artificial intelligence cannot be reduced to technical instruction alone. Instead, it requires a holistic, critical, and ethically grounded approach that situates AI within broader social, political, and economic contexts. AI systems are not neutral tools: they are shaped by human decisions, embedded values, and institutional power structures. As such, they play an active role in shaping inequalities, opportunities, and forms of participation in society.
To teach AI critically is therefore to move beyond understanding how systems function, toward examining how they affect people, redistribute power, and influence life chances. This involves recognising that data systems are deeply implicated in the reproduction of inequality. From predictive policing to algorithmic recruitment, AI can encode and amplify structural injustices, disproportionately affecting marginalised communities. These dynamics are further reinforced by systems of surveillance and behavioural manipulation, which shape how individuals are monitored, categorised, and governed.
Within this landscape, education has a crucial role in fostering data agency. Learners must not remain passive subjects of data systems; rather, they should be supported to become informed and critical participants who can understand, question, and challenge how data is used. This shift from passivity to agency is fundamental to democratic participation in a data-driven society. Finally, these aims cannot be achieved through isolated teaching efforts. Data literacy, ethical reasoning, and critical AI understanding must be embedded across the curriculum, integrating interdisciplinary perspectives and aligning with broader digital competence frameworks. In doing so, education can become transformative—equipping students not only with knowledge, but with the capacity to engage critically, act responsibly, and advocate for more just and equitable data practices.
flowchart TD
A[Teaching AI & Data] --> B[Technical Understanding]
A --> C[Critical Engagement]
A --> D[Ethical Reflection]
C --> E[Examine Bias in Data]
C --> F[Analyse Social Impact]
C --> G[Interrogate Power Structures]
D --> H[Data Ethics]
D --> I[AI Ethics Principles]
D --> J[Privacy & Rights]
E --> K[Data Inequality Awareness]
F --> K
G --> K
K --> L1[Racism & Bias]
K --> L2[Gender Inequality]
K --> L3[Socioeconomic Discrimination]
K --> L4[Surveillance & Manipulation]
K --> M[Critical Consciousness]
M --> N[Data Agency]
N --> O1[Understand Data Use]
N --> O2[Make Informed Choices]
N --> O3[Challenge Systems]
N --> P[Active Data Citizenship]
P --> Q1[Participate in Democracy]
P --> Q2[Advocate for Justice]
P --> Q3[Reshape Data Practices]
%% Curriculum integration
A --> R[Embedded Across Curriculum]
R --> S[Interdisciplinary Teaching]
R --> T[Ethics in Assessment]
R --> U[Digital Competence Frameworks]
%% Styling (pastel + black text)
classDef pastel fill:#fff0f5,stroke:#555,color:#000;
classDef pastel2 fill:#e6f7ff,stroke:#555,color:#000;
classDef pastel3 fill:#f0fff0,stroke:#555,color:#000;
class A,R,S,T,U pastel;
class B,C,D,E,F,G pastel2;
class H,I,J,K,L1,L2,L3,L4 pastel;
class M,N,O1,O2,O3,P,Q1,Q2,Q3 pastel3;
To support this integrated approach, teaching should focus on:
- Critical engagement with AI systems
- How bias is embedded in datasets and algorithmic design
- How AI-driven decisions affect different social groups in uneven ways
- How power is distributed and exercised through data and algorithmic systems
Understanding data, inequality, and social justice
The role of data in reproducing racism and structural bias
- Gender inequalities embedded in algorithmic decision-making
- Socioeconomic discrimination through automated systems (e.g., welfare, education, insurance)
- The expansion of surveillance and its implications for autonomy and rights
- The use of data and AI for manipulation, particularly in political and social contexts
Fostering data agency
- Understanding how personal and collective data shapes everyday life
- Developing the ability to make informed decisions about data sharing and use
- Building confidence to question, challenge, and contest harmful data practices
Embedding across academic practice
- Integrating data ethics and AI literacy across disciplines
- Designing interdisciplinary learning experiences
- Embedding ethical reflection into research and assessment
- Aligning teaching with digital and data competence frameworks
Designing transformative learning outcomes
- Enabling students to critically engage with data-driven systems
- Supporting participation in democratic and policy debates حول data and AI
- Encouraging advocacy for ethical, inclusive, and socially just data practices
The ultimate aim of this approach is to ensure that learners do not simply understand AI systems, but are equipped to:
- Critically interrogate them
- Challenge their assumptions and impacts
- Actively reshape data practices toward more just futures