Introduction to the Raspberry Pi 5 and AI HAT+ Setup

This month you move from microcontrollers to a full microcomputer: the Raspberry Pi 5, boosted by the AI HAT+ for real-time artificial intelligence. This first lesson has no coding; it explains why the Pi is a different class of machine, what the AI HAT+ adds, and how to set everything up so you are ready to build a smart security camera, run local AI models, and create a voice assistant later this month.

Microcontroller versus microcomputer

Everything so far ran on a microcontroller, a single chip built for embedded control: it has a processor, a bit of memory, and input/output pins, and it runs one program over and over to do one job well. The Raspberry Pi 5 is a different kind of machine, a full microcomputer with a processor, RAM, storage, and peripheral ports for keyboards, mice, cameras, and displays. It runs a complete operating system, so it can multitask, browse the web, play video, and run AI models. Where a microcontroller is focused, the Pi is versatile.

What the Raspberry Pi 5 offers

The Pi 5 has a quad-core processor, 8GB of RAM, and runs Raspberry Pi OS, a Linux-based operating system. It has built-in Wi-Fi, Ethernet, and Bluetooth, plus a bank of GPIO pins so it can still talk to motors, sensors, and displays like a microcontroller when needed. It supports full HD video and a camera input, and it runs many languages including Python, which you will use this month. In short, it brings desktop-class capability into a board that fits in your hand.

The AI HAT+

The AI HAT+ plugs onto the Pi like a graphics card plugs into a computer, adding dedicated AI horsepower, about 13 TOPS, meaning trillions of operations per second. That performance is what makes real-time AI practical on a small board: it can run object detection, image classification, and vision models like YOLO or MobileNet in milliseconds rather than seconds. Without it, the Pi could still run AI, just far more slowly; with it, live camera analysis becomes smooth.

What you will build this month

Because the Pi is a real computer, the projects step up accordingly. You will build a smart security camera, run local AI models directly on the device (no cloud required), perform real-time image recognition where a live camera labels what it sees, and finish by building your own voice-driven AI assistant, your own Siri or Alexa. Today is only setup, because getting the Pi imaged and booting takes a little time and everything else depends on it.

Working through it

Install the heat sink. Attach the heat sink to the Pi 5. Like any computer, its processor gets warm under load, and the heat sink keeps it cool during AI work.

Flash the SD card. On your separate computer, install the Raspberry Pi Imager from raspberrypi.com/software, insert the 32GB SD card, choose Raspberry Pi OS 64-bit (needed to use all 8GB of RAM), select the card, and write.

Preconfigure in the Imager (recommended). In the Imager's advanced options, set a password, enable SSH, and enter your Wi-Fi ahead of time. Doing this now saves setup steps later on the Pi itself.

Assemble the hardware. Insert the flashed SD card, connect the AI HAT+ to the GPIO header, attach the touchscreen, and plug in power. Add a keyboard and mouse, or use the touchscreen.

Boot to the desktop. Power on and confirm the Pi boots into the Raspberry Pi OS desktop. That is the finish line for today and the foundation for everything that follows.

Common mistakes and troubleshooting

Flashing the 32-bit OS. Choose the 64-bit Raspberry Pi OS so the Pi can use all 8GB of RAM, which AI work needs.

Skipping the heat sink. Install it before heavy use; the processor throttles or overheats without cooling during AI workloads.

The Pi does not join Wi-Fi on first boot. Set the network in the Imager's advanced options before flashing, or configure it from the desktop after booting.

The AI HAT+ is not seated on the GPIO header. Press it firmly and fully onto the header so all pins connect before powering on.

Next: Next you will run real AI on sample images and video to see what your new setup can do.