CSTA Standards Mapping
Her AI Studio curriculum → 2026 CSTA PK–12 Computer Science Standards (Grades 9–12)
Every lesson in this curriculum is mapped below to the 2026 CSTA standards it addresses. The AI-aware structure of the 2026 revision suits us well: AI content is woven throughout the five foundational concepts (Algorithms & Design, Programming, Data & Analysis, Systems & Security, Computing & Society) and has its own High School specialty strand (Artificial Intelligence / AIN) with two tiers.
How to read the standard IDs
| Prefix | Meaning | Who it's for |
|---|---|---|
HS-* | High School foundational standard | All students in grades 9–12 |
S1-* | Specialty I (first focused experience in a domain) | HS students taking an AI/physical-computing pathway |
S2-* | Specialty II (advanced study) | HS students in deeper or capstone coursework |
Concepts: ALG = Algorithms & Algorithms, PRO = Programming, DAT = Data & Analysis, SYS = Systems & Security, SOC = Computing & Society, AIN = Artificial Intelligence, PHY = Physical Computing.
CSTA source: 2026 CSTA PK–12 Computer Science Standards (CC BY-NC-SA 4.0).
Week 1 — Understanding AI: Who Built It, Who It's For, and Why You Belong Here
Beginner course. Students define what an AI model is, learn the history of AI's women builders, confront bias as a mirror, and practice skepticism as a skill (AI Audit Card, personal statement).
Core standards (plan your assessment evidence around these):
| Standard | Title (abbreviated) | Why it aligns |
|---|---|---|
S1-AIN-DD-01 | Analyze AI systems to differentiate the types of problems they address | The Two Truths and a Lie warm-up and the "generative vs. discriminative" distinction (chatbots vs. image classifiers). |
S1-AIN-HR-09 | Analyze the potential biases and limitations of AI systems | Your "bias is a mirror" lesson and the Gender Shades discussion — this standard's own example is a facial-recognition performance gap across demographic groups. |
HS-ALG-PS-05 | Evaluate AI-generated output to assess bias, accuracy, and potential harms | The AI Audit Card's "does it work for everyone?" questioning is a structured version of this standard. |
HS-ALG-IM-11 | Justify values embedded in the design of an algorithmic system | "Who benefits? who built it? what data trained it?" — every design choice prioritizes some value (efficiency, privacy, fairness). |
HS-SOC-HI-38 | Analyze a technology's historical trajectory and its societal/environmental links | The three-women-in-AI history flipchart and "why your perspective matters." |
Supporting standards:
HS-ALG-PS-04— Describe deterministic vs. probabilistic algorithms (the "model predicts the next word" mini-lecture; ML is statistical, not rule-following)HS-ALG-ML-07— Evaluate training data by source, quality, and representativeness (audit-card question: "What data do you think it was trained on?")HS-ALG-IM-10— Evaluate ethical impacts and bias of rule-based vs. data-driven algorithms (two sources of bias)HS-SOC-HU-43— Evaluate how human choices in designing/deploying/regulating tech create risks and long-term impacts (the "demands accountability" skeptic move)HS-SOC-HU-44— Debate human vs. AI intelligence, consciousness, and responsibility (warm-up "AI doesn't think like a human")HS-SOC-ET-41— Evaluate societal/environmental impacts of emerging tech, incl. inequities (who might it work less well for?)S1-AIN-PP-12— Analyze AI tool experiences across diverse user backgrounds (accompaniment, language, geography)S1-AIN-PP-13— Assess how unauthorized data/AI-copyright issues affects training practice (Miyazaki reading; Kenya labor reading)S1-AIN-PP-14— Evaluate how AI ethics has changed over time (rule-based → model-based evolution)HS-SOC-CE-45/HS-SOC-CE-46— Diverse teams and your place in this field (the "you belong here" charge)
Week 2 — Taking Control: Training Data and Local AI
Beginner · Students contrast cloud vs. local AI, explore privacy and data sovereignty, install and run an open-source model (Ollama), and train a custom image classifier with the My Room app (collect → train → infer).
Core standards:
| Standard | Title (abbreviated) | Why it aligns |
|---|---|---|
HS-ALGS-ML-08 | Develop a machine learning model for a chosen task using appropriate data and tools | The My Room training loop — every student trains their own classifier. |
HS-ALG-ML-07 | Evaluate training data: source, quality, representativeness, bias, privacy | Choosing and evaluating your own collected classes/examples for "My Room." |
| `S1-AIN-DD-03Create an application using a prebuilt supervised learning model to classify or predict | My Room uses prebuilt MobileNet features — the standard's own example. | |
S1-AIN-DS-07 | Apply data acquisition, cleaning, and transformation techniques for AI analysis | Your data-prep steps before training (naming classes, balancing, splitting). |
HS-DAT-IM-27 | Evaluate societal, environmental, ethical implications of large-scale data | "The cloud is just someone else's computer" — this standard literally names data sovereignty and data-center impact. |
Supporting standards:
S1-AIN-DD-02— Modify training data to improve fairness/accuracy (rebalancing with more objects / more classes)S1-AIN-DS-06— Examine data flow through a neural network (the "how the classifier works" sketch)HS-ALGS-ML-06— Justify model choice (image classifier vs LLM vs rule) (your local-model comparison table)HS-DAT-DC-23/HS-DAT-DC-24— Clean/inspect data consistency (data hygiene before training)HS-SYS-SE-31/HS-SYS-SE-32— Security measures/tradeoffs and breach causes & impacts (privacy trade-off of cloud vs. local)HS-SYS-NT-35— Internet as network-of-networks; client/server model (drawing where your query goes)HS-SYS-HW-29— Operating system roles managing hardware (running Ollama locally)HS-SOC-HI-39— Propose policy for ethical innovation (GDPR/data-sovereignty discussion)HS-DAT-IM-28— Debate data-use policy efficacy (data sovereignty debate)S1-AIN-HR-08— Plan safeguards protecting well-being & privacy (guardrails for your local model)S1-AIN-HR-10— Analyze environmental impacts of widespread AI adoption (edge vs. data-center energy)S1-AIN-PP-11— Integrate a prebuilt AI agent (an LLM) into an app (Ollama powering journal entries / code)
Week 3 — Hello, a network of computing systems: From PC Control to Networked AI
Intermediate · hardware. Students set up an Arduino IoT board, write their first sketches setup()/loop(), control LED array, read mouse input, and connect board to local LLM over WiFi/HTTP via Python bridge.
Core standards:
| Standard | Title (abbreviated) | Why the alignment |
|---|---|---|
S1-PHY-HC-01 | Construct a computing circuit to power & control a physical device | The Uno Q + USB/mouse-LED-array circuit build. |
S1-PHY-IO-02 | Integrate sensors, actuators, peripherals into a physical system | Mouse input, LED matrix output, later camera input. |
S1-PHY-DD-03 | Develop software iteratively to control a physical device | Every Arduino setup() / loop() sketch and its debugging runs. |
S1-PHY-PP-08 | Evaluate security implications of a physical computing system | Running at 0.0.0.0, exposing a local server, and the "data stays on your network" discussion. |
HS-SYS-NT-34 | Diagram a network of computing components (hardware + software) | The Uno Q → WiFi → Python server → Ollama pipeline diagram. |
HS-SYS-NT-35 | Analyze the internet's client/server model and network-of-networks | HTTP request/response over LAN — the "where did my query go" follow-through. |
**Supporting:
HS-PRO-PD-12— Modular programs via#includelibraries; lessons use external Arduino/Python modules.HS-PRO-PD-13— Use documentation, libraries, APIs (Arduino LED library, Flask,ollamasubprocess)S1-AIN-PP-11— Integrate a prebuilt AI agent into an application (the LLM behind the Python server)HS-SYS-SE-33— Formulate a solution to a security flaw (discussing the open server/trust boundaries)HS-PRO-RD-18— Evaluate AI-generated code for accuracy/reliability (AS-demanding; assess code)S1-PHY-CI-05— Implement IoT communication with basic protocols (WiFi/HTTP) (used again in Week 4)
Week 4 — Can You See Me Now? Adding Computer Vision to Your Arduino
Intermediate · vision + LJ. Students add a camera, stream video over WiFi, classify with MobileNet on the laptop, chain vision → LLM prompt → display + LED feedback, and reflect on confidence scores and multi-modal pipelines.
Core standards:
| Standard | Title (abbreviated) | Why the alignment |
|---|---|---|
S1-AIN-DD-03 | Create an app using a pretrained supervised learning model to classify | MobileNet image classifier on the laptop — direct example used by the standard itself. |
S1-AIN-DD-04 | Compare data representations & constraints (image vs. text) | Same system handles image data (MobileNet) and text data (the LLM) — compare both streams. |
S1-AIN-DS-06 | Examine how data flows through a neural network | Teacher-era notes: labels, weights, layer-by-layer flow, output logits → confidence. |
S1-PHY-IO-02 | Integrate sensors/peripherals (camera) into a system | The Arducam OV2640 over SPI, image capture. |
S1-PHY-CI-04 | Use IoT devices to collect & transmit sensor data locally (device-to-gateway) | Camera on Uno, WiFi to laptop — no cloud involved, exactly as CSTA ran. |
S1-PHY-PP-08 | Evaluate security implications of a physical system | Discussion of who/ідікс network; keeping personal data locally. |
Supporting standards:
HS-ALG-ML-07— Evaluate the model's training influence incl. representing limitation (MobileNet's 1000 classes, gaps) — "Notes which the model gets right & struggles → why"HS-DAT-DI-26— How to app update An video: "visualization/simulation limitations" maps to confidence-score critiqueHS-DAT-DC-23/24— Handle data fidelity in image transfer (format, size, JPEG)HS-SYS-NT-34— reiterate the 5-stage pipeline diagram from Week 3, now with image dataHS-PRO-PD-13— libraries/APIs (Requests, PIL, TensorFlow, Keras)S1-AIN-DD-05— Evaluate whether AI or non-AI would be more appropriate (e.g., rule-based color detection vs. an image classifier)HS-SOC-ET-41— Debrief: for whom does a camera system work better/worse?
Week 5 — Creating a Fully Local Computer
Advanced · Alone. Work toward a server-independent, fully offline local build. (Draft week.)
| Standard | Title (abbreviated) | Why aligns |
|---|---|---|
S2-PHY-CI-04 | Develop an IoT system to manage local data collection/transmission | Coordination of local devices w/o 'internet*-medium' — offline logic. |
S2-PHY-DD-03 | Develop an application extending functionality and user engagement | Level up the desk companion into a standalone tool. |
HS-SYS-HW-30 | Demonstrate a device's capabilities & limitations | Running Ollama/SBC offline — the limits are the point. |
HS-ALG-IM-09 | Design with human-centered design principles | Who will use this device? Accessibility, power, portability. |
HS-PRO-TR-19 / HS-PRO-TR-20 | Evaluate against specs & responsible values; iterate with feedback | Usability tuning toward the final build. |
Week 6 — My Cyberdeck: The Final Build
Advanced · capstone build.
| Standard | Title (abbreviated) | Why aligns |
|---|---|---|
S2-PHY-PP-08 | Develop a physical computing solution for diverse users using an engineering design process | The cyberdeck is your standard's reference implementation (sketch → plan → build → test → refine). |
S2-PHY-PP-06 | Apply a project management methodology to a collaborative build | Track scope: prototyping boards, wiring, programming, user testing. |
S2-PHY-IO-02 | Design a closed-loop feedback that behaves as you intend | Cyberdeck power/portability/uptime under load. |
HS-ALG-IM-09 | Human-centered design | Portability, accessibility, personalization — "your own voice." |
HS-SOC-ET-42 | Design a conceptual solution to a real-world problem with an emerging technology | Framing the deck as a device that serves a purpose you choose. |
HS-PRO-TR-19/20 | Evaluate and refine the build | Final acceptance tests + post-build reflections. |
Capstone — My Capstone: Present and Teach
Capstone. Students explain, showcase, and paginate their cyberdeck and its AI pipeline, teaching others what they built and why it matters.
| Standard | Title (abbreviated) | Why aligns |
|---|---|---|
S2-AIN-PP-09 | Develop professional communication artifacts translating technical AI results for diverse audiences | The capstone talk/demo/site tailored to peers vs. judges vs. parents. |
HS-SOC-CE-46 | Connect the computing you learned to personal goals and interests | The "I can teach this" final reflection. |
HS-SOC-CE-45 | Elaborate how diverse teams use computing to solve problems | The dean for future cohorts / the "it was built by people like us" moment. |
HS-PRO-PD-14 | Attribution & intellectual property | A proper credits page for code, libraries, icons. |
HS-PRO-TR-19 | Evaluate the build against specs & responsible values | The final self-audit of your own project. |
HS-SOC-ET-41 | Evaluate broader impacts | "Who would benefit most from a tool like this?" reflection. |
Quick reference — one-line coverage
| Week / Block | Go-to standards |
|---|---|
| W1 — AI identity, history, bias, skepticism | S1-AIN-HR-09, HS-ALG-PS-05, HS-ALG-IM-11, HS-SOC-HI-38 |
| W2 — Training data, local AI, privacy | HS-ALG-ML-08, S1-AIN-DD-03, HS-DAT-IM-27, HS-ALG-ML-07 |
| W3 — Arduino & networks | S1-PHY-HC-01, S1-PHY-IO-02, S1-PHY-DD-03, HS-SYS-NT-34/35 |
| W4 — Computer vision & multi-modal | S1-AIN-DD-03, S1-AIN-DS-06, S1-PHY-CI-04, HS-DAT-DI-26 |
| W5–6 — The cyberdeck | S2-PHY-PP-08, S2-PHY-CI-04, HS-PRO-TR-19/20 |
| Capstone — Present & teach | S2-AIN-PP-09, HS-SOC-CE-46/45, HS-PRO-PD-14 |
Notes on this mapping
- Specialty vs. foundational: The AI (
AIN) and Physical Computing (PHY) sites are high-school specialty standards; theHS-*items are the foundational band every 9–12 student should reach. Combined, they let you show both "for all" and "for pathway" coverage. - Practices & dispositions: Most aligned standards carry the ESR (Ethics & Social Responsibility) and HCD (Human-Centered Design) practices, which describe how students engage — that is why we grouped them as the natural homeowners of your pedagogy.
- The CSTA names AIN standards as tiers (Specialty I → II) so you can place the semester goal (e.g.,
S2-PHY-PP-08) while still meeting the beginner expectation (S1-PHY-HC-01).
