Prof. Modestino | Howard University

Week 4 Feb. 2, 4, and 6

5 min read

Monday Slide Deck

Race After Technology: Chapters 1–4

What is the “New Jim Code”?

“The employment of new technologies that reflect and reproduce existing hierarchies… even as they are often promoted and perceived as more objective or progressive.”

— Ruha Benjamin (p. 5)

The Reality Check:

  • It isn’t just about “racist people.”

  • It’s about racist systems being automated.

  • Tech launders bias: it makes it look like math instead of prejudice.

Chapter 1: Engineered Inequity

The Myth of Neutrality

  • Core Argument: We treat tech as a fix for human messiness and inefficiency

  • But we Have to Pause: If you build a high-speed train that only stops in wealthy neighborhoods, the train isn’t neutral—the map it follows is biased.

  • Discussion: Why do we trust a computer’s “score” more than a human’s story?

Chapter 2: The Logic of the “Glitch”

A Glitch is a Signal, Not a Mistake

  • Benjamin’s Argument: A glitch reveals who the system was designed for and who was ignored (p. 77).

  • The Default: Who is the “Standard User” in the designer’s mind?

  • Reality Check: * If a soap dispenser doesn’t “see” dark skin, it’s not a “broken” sensor.

  * It is a sensor performing exactly as it was taught—on a white “default” body.

Chapter 3: Coded Exposure

The Trap of “Being Seen”

  • The Shirley Card: Historically, film was calibrated using a white woman’s skin tone as the “standard” (Benjamin, 2019, p. 112).

  • The Reality Check: For many, “inclusion” is not the goal—it is a trap.

  • The Counter-Argument: If we “fix” facial recognition to see Black faces perfectly, does that help Black people, or does it just make it easier for the state to monitor them?

  • Key Concept: Visibility Justice.

Chapter 4: Technological Benevolence

  • Technobenevolence: Tech that claims to “help” or “diversify” while actually extracting data or controlling people.

  • Workday & Howard: Howard University uses Workday; our leadership sits on their board.

  • Reality Check: Workday markets AI as “removing human bias.” But if the AI is trained on “standard” corporate data, it may treat some degrees as outliers or risks.

  • Discussion: Is bias-free tech possible if the world it’s built in is still unequal?

Reading Code:

 
# A plausible logic for an automated resume scan (ATS)
 
def evaluate_candidate(data):
 
    # 1. Geographic Filter (Legal, but reflects Redlining)
 
    if data['miles_from_office'] > 20:
 
        return "Low Priority"
 
  
 
    # 2. Employment Gap (The Default of constant work)
 
    # Punishes caregivers and those with economic instability
 
    if data['unemployed_more_than_6_months'] == True:
 
        return "Reject"
 
  
 
    # 3. Pedigree Filter (Reflects class privilege)
 
    if data['college'] not in target_school_list:
 
        score -= 20

Discussion: Authorship & Accountability

  1. Authorship: If Workday’s “Predictive Analytics” flags a student worker as a retention risk based on their zip code, who is the author of that decision?

  2. The Default: How do we interrogate the default settings of the software we use every day on campus?

  3. Writing as Power: If coding is writing, then critiquing code is a form of essential literacy for the 21st century

Feb 4:

Introduction to Midterm Assignment

9. Midterm Assignment

Discussion of Center for Democracy and Technology Report on AI in Schools

Link Hand in Hand: Schools’ Embrace of AI Connected to Increased Risks to Students

Feb 6th Writing Lab.

Your Algorithmic Proxy

Overview

Ruha Benjamin argues that algorithms are not neutral; they are “coded reflections” of social priorities. In this lab, you will reflect on your own digital algorithmic environment. Your goal is to move from a consumer of content to an observer of the system that predicts you.

Step 1: The Prediction Log (~15 Minutes)

Open your most-used algorithmic feed (TikTok, Instagram, YouTube, or Spotify). Document the first five items presented to you. For each item, record:

  1. The Content: A brief description of the video/post/song.

  2. What data breadcrumb did you leave behind that you think led the machine here? (e.g., a specific search, a long pause on a previous video, your GPS location).

Step 2: Defining your “Algorithmic Proxy” (15 Minutes)

Benjamin discusses how data systems use “proxies” to categorize us. Based only on those five items, answer the following:

  • If a stranger saw only this feed, what “User Category” would they assign you? (Give this category a name, e.g., “The Doomsday Prepper,” “The Fast-Fashion Consumer,” “The Corporate Striver”).

  • Identify one significant part of your identity, hobby, or value system that is entirely missing from this feed. Why might this part of you be invisible to the algorithm’s profit model?

  • Look at an item in your feed that feels wrong or like a stereotype. Is it a mistake, or does it reveal an assumption the developers made?

Submission Requirements

Submit a 250-word summary including:

  1. Your Data Log: The 5 items and your proxy category name.
  2. Analysis: Describe the part of you the machine misses or gets wrong and why that silence or assumption in the data matters.

Graph View

Backlinks