CMPSC 303 · Fall 2026

Artificial Intelligence

Dr. Janyl Jumadinova · Tuesday and Thursday 11:00 to 12:15 · Lab Tuesday 2:30 to 4:00 · Alden 101

Seventy years in five minutes

It started as a question about behavior

• 1950. Turing asks "can machines think?", decides the question is hopeless, and replaces it with a test you can actually run.
• 1956. A summer workshop at Dartmouth gives the field its name. The proposal asks for two months and ten people.
• They were wrong about the timeline. So has nearly everyone since.

Seventy years in five minutes

Two answers, and a long argument

• Write the rules down. Logic, search, knowledge encoded by hand. Chess, theorem proving, expert systems.
• Learn them from data. Statistics and neural networks finding patterns nobody wrote out.
• The field has swung between these twice, and each side has declared victory more than once.

This course does both, in that order. That is not an accident.

Seventy years in five minutes

Boom, winter, repeat

• 1960s. Confident predictions of human-level machines within a generation.
• 1974 and 1987. Funding collapses twice. People call them the AI winters.
• 1997 Deep Blue. 2012 AlexNet. 2016 AlphaGo. 2017 transformers. 2022 ChatGPT.
• Every one of those was called the beginning of everything. Some of them were.

Seventy years in five minutes

Which brings us to now

• This is the largest boom the field has ever had.
• During the last four, the people in the room could not reliably tell progress from hype either.
• Being able to tell the difference is most of what this course is for.
• That requires knowing how the things actually work, which is where we are going to start.

Everything lives here

Schedule, syllabus, and these slides. Bookmark it now.

areweagentsyet.com

I built this with Claude. I created agentic workflow. It wrote most of the code. I made the decisions and reviewed every page, and the mistakes are mine.

How this class works

• Challenges: nine coding assignments, one per concept. Each has a written reflection and a short verbal check on your code.
• Briefings: five times this term, your group gathers material on a theme and reports back to the class.
• Projects: three larger builds, each with a presentation, a code review, and a written reflection.
• Concept checkpoints: three short individual check-ins.

What you need

You already have these from CMPSC 100 and 101. Check that they still work before Week 2.

• uv, which manages Python for you.
• A working git setup.
• VS Code.
• Accounts on GitHub and Discord.
• Sign up for the GitHub Student Developer Pack if you have not already.

Survey

How does this room think about AI?

Nine questions. Anonymous. No wrong answers.

forms.gle/gvDSGKXh9E6DYAVw8

• Take about five minutes.
• Then we look at the results together.

Activity

Stand where you actually are

One end of the room is strongly agree. The other end is strongly disagree. No middle huddling.

• "AI should be allowed on all coursework."
• "I use AI more than most people my age."
• "I am more excited than worried about AI."
• "AI will make my degree less valuable."
• "Netflix recommendations count as AI."

After each one: two people from different spots say why they are standing there.

Nationally

How students answered

Put these next to our numbers on screen.

• 88% of students use AI in their learning, up 16 points in a single year.
• 57% use it weekly for coursework. 21% use it every day.
• Students in technology and engineering use it most of all.

Digital Education Council, 2026 (35 countries) · Lumina Foundation and Gallup, 2026 State of Higher Education

Nationally

How people feel

• 52% of Americans are more concerned than excited about AI. Only 9% are more excited.
• In 2021, the concerned number was 37%.
• 55% of adults under 30 are more concerned than excited. That is the first time it has been a majority.
• 71% expect AI to mean fewer jobs within twenty years.

Pew Research Center, June 2026 survey · under-30 figure published August 18, 2026

What it costs to run

The part that does not fit in a chat window

• Electricity use by AI data centers grew about 50% in 2025.
• Data center demand overall is on track to roughly double by 2030.
• Cooling all of it takes water, and the companies involved disclose very little about how much.

International Energy Agency, Energy and AI

Nobody has asked Allegheny

The country gets surveyed constantly. This campus does not.

• Do people here look anything like those national numbers?
• Do people outside computer science define AI the way we just did?
• Allegheny is discussing right now what it does about AI, from plagiarism to accessibility.

Before Thursday

Bring one story

Nothing to submit. Can work in pairs.

• Find one recent story about AI that genuinely surprised or unsettled you.
• Recent means the last three months.
• Bring the link. "I heard that AI..." does not count.
• Be ready to say in thirty seconds what happened and why it got you.