D790 is a course in WGU's M.S. Software Engineering, AI Engineering specialization.
WGU describes it as showing how AI systems are designed to align with human behavior and values, characterizing human-centered design for AI, analyzing ethical implications for individuals and society, and explaining how to build in human oversight and control.
It is a software engineering course, not a philosophy seminar, which surprises some students. Identifying a harm is only half the task; the course expects you to turn it into design decisions such as who can override the system, what is logged and how impact is monitored over time. The guide lists three competency units.
Course at a Glance
| Item | Details |
|---|---|
| University | Western Governors University (WGU) |
| Course code | D790 |
| Level | Graduate (3 competency units in the guide) |
| Program | M.S. Software Engineering, AI Engineering |
| Subject area | Human-centered AI, ethics and oversight |
| Typical work | An analysis of an AI system with proposed oversight and monitoring |
What D790 Covers
| Competency (summarized) | What it means in practice |
|---|---|
| Characterize human-centered design and apply it to AI | Design around users' needs, context and values |
| Analyze ethical implications for individuals and society | Identify who is affected and how |
| Explain mechanisms for human oversight and control | Decide where humans review, override or shut down |
| Describe how impact will be monitored and the system improved | Plan feedback, measurement and iteration |
Key Concepts Explained
Human-Centered Design for AI
Human-centered design starts with the people who use or are affected by a system. For AI it adds questions about transparency (can users understand why?), agency (can they contest or opt out?) and trust (is confidence communicated honestly?).
Example: A resume-screening tool should tell recruiters why a candidate was ranked low and let them review rejected applications, instead of silently filtering people out.
Ethical Analysis
Common themes are bias in data, privacy, accountability, safety and effects on work. A useful analysis names the stakeholders, the possible harm, how likely and severe it is, and who is responsible.
Example: If a hiring model was trained on past decisions that favored one group, it may reproduce that pattern. The analysis names affected applicants, explains the cause (biased training data) and proposes a fairness check across groups.
Human Oversight Models
Oversight can be human-in-the-loop (a person approves each decision), human-on-the-loop (a person monitors and can intervene) or human-in-command (a person decides when the system is used). The right level rises with the stakes.
Typical Assignments and How to Approach Them
| Task element | What it tests | How to approach it |
|---|---|---|
| Stakeholder and impact analysis | Ethical reasoning | Table of stakeholder, possible harm, likelihood and severity |
| Human-centered design review | Applying principles | Assess transparency, agency and trust for the named system |
| Oversight design | Concrete mechanisms | Choose an oversight level and say who acts, when and with what tools |
| Monitoring plan | Continuing alignment | List measures, review cadence and triggers for change |
From Concern to Control
The bridge between analysis and design is a simple table. Each row takes one concern from your ethical analysis and turns it into a mechanism, an owner and a measure.
| Concern | Mechanism | Owner and measure |
|---|---|---|
| Biased outcomes | Regular fairness audit across groups | Data lead; gap between groups |
| Opaque decisions | Plain-language explanation shown to users | Product owner; user understanding feedback |
| Over-reliance | Human review of high-stakes cases | Operations lead; override rate |
| Privacy risk | Data minimization and retention limits | Privacy officer; data held versus needed |
Readiness Checklist Before You Submit
Use this list to confirm the analysis has become design.
| Check | What to confirm |
|---|---|
| Stakeholders named | Users and affected groups are identified |
| Harms specific | Each concern describes who is affected and how |
| Mechanisms concrete | Each concern has a control, an owner and a measure |
| Oversight level chosen | You say where humans review, override or stop the system |
| Monitoring planned | Impact will be measured and the system improved over time |
Where Students Get Stuck
- Abstract ethics. Name specific harms and specific mechanisms, not general principles.
- Oversight with no owner. A control needs a person or role responsible for it.
- One-time review. Monitoring should continue after launch.
- Ignoring users. Include how people affected can question or appeal outcomes.
Study Tips for D790
- Keep a standard stakeholder table you can adapt to any system.
- Learn the three oversight models and one example of each.
- Use concrete numbers for monitoring, such as review cadence, even if illustrative.
- Keep your tone analytical; evaluators look for reasoned judgment.
How We Help with D790
Send the scenario, the rubric and any draft. A writer with a technology ethics or software background can prepare a worked model for study, explain oversight models, or review your analysis for gaps between concern and control. Our master's assignment help guide explains the process.
GradeEssays is independent of Western Governors University. Model work is for study and reference; what you submit must be your own under WGU's academic integrity policy, and no outcome is promised. Orders are written from scratch with free revisions within the scope of your original request, and a full refund if late.
Make Sense of Your D790 Task
Share the scenario, rubric and draft. A writer prepares a worked model you can study.
Start My D790 HelpFree revisions · Full refund if late · Written from scratch for your order
Frequently Asked Questions
WGU's guide lists it in the M.S. Software Engineering, AI Engineering plan.
Three, according to the guide.
The listed competencies are analytical and design-focused. Check your task for any technical deliverable.
A design where a person reviews or approves the system's decisions before they take effect.
Define measures, a review schedule, responsible roles and the actions triggered when results drift.
It follows Applied Machine Learning for Business Solutions in the usual order, before Integrating AI with Modern Software Applications.