Completing a DIY project used to mean going to the hardware store or hobby store and digging out tools from your garage. Today, a new onslaught of generative AI products like ChatGPT and easy-to-use training platforms have added another dimension to these builds, making them not only satisfying to create, but realistic to run.
These are three DIY projects that integrate AI into their designs to create moving art, easy-to-use health detectors, and an escape from boring lectures.
Kellyn Carolyn Chan and Ryan Mather Independent designers work together to design “Poetic Camera”. It's a deceptively simple camera that uses a Raspberry Pi and ChatGPT to take photos and output an AI interpretation of the images in the form of original poetry.Chan and Mother say: popular mechanisms By using ChatGPT with Poetry Camera, we were able to not only bring our projects to life faster, but also give them an essential human touch.
edward aguilar I was also considering adding a human element to my project when building with ChatGPT. Beulah bot. Currently, Aguilar is the co-founder and CEO of a human-level transcription startup. echo lab. Before taking on the role, he created his BuellerBot. ferris bueller's day off Roll Call Scene – Free your students from boring video lectures by giving thoughtful answers when AI is called on their behalf.
These AI projects also serve more practical purposes, such as: Diana Karipina's e-health project includes a “fatigue detector” based on monitoring the user's seat height and eye opening. Khalipina is a freelance web developer based in Paris.
Will AI become the new essential element for DIY projects? According to these designers, it can be.
What are the benefits of AI in DIY projects?
For Zhang, building Poetry Camera might not have been possible without the help of ChatGPT.
“AI was important not only to make the camera itself work, but also to lower the barrier to starting the project,” says Zhang. “This was a re-introduction to programming for me after nearly 10 years away. Thanks to ChatGPT, in just January he was able to go from knowing nothing to writing working code in four days. Now.”
Zhang says that by incorporating AI into Poetry Camera from the beginning, they were able to give the project essential impetus. Aguilar says this instant gratification from idea to prototype has also benefited Bueller Bot.
“This would not have been possible without this technology being so easily accessible,” Aguilar said. popular mechanisms. “There are a lot of ideas that are just hobbyists that you want to consider but don't want to spend months on. With AI, he can do in a year almost everything that a team of a few dozen people can do in a year. .Weekend.
Accessibility was also a key benefit for Karipina. Google's Teachable Machine To train her fatigue sensor.
“[I] We had a dream to create an e-health project that would analyze health conditions based on video analysis…but it was difficult from a hardware point of view,” says Karipina. popular mechanisms. “Then I learned about Google's Teachable Machine and was intrigued by its real-time video analysis opportunities. It was quick, convenient, and didn't require expensive hardware.”
Integrating AI into projects can also be important for project creativity, Mather says.
“We really need AI for this project, because without it, the poems would just look very crude and random,” Mather says. “When people know that the AI 'saw' something really specific or personal in a poem, there is a wonderful magical moment when people light up with childlike joy and playfulness.”
Human connection and creativity are core principles at Poetry Camera, and the creative use of AI can help rather than hinder, Zhang says.
“In my opinion, all AI art projects are about restraint and perspective. There’s so much that can be done with this tool that the creator’s taste and intentional decisions matter. how It’s more important than ever to embrace AI,” she says. Decisions like never reading a poem aloud, never printing a source photo, or open sourcing a DIY procedure.
“All of these decisions are about fostering human connection and human creativity, and that's really our end goal,” Chan says.
Overcoming the challenges of using AI
Although the integration of AI into the project went well, Zhang and Mather say they still face challenges that plague most DIY users, especially hardware vulnerabilities.
“I [learned] “Hobby electronics are much less durable than consumer electronics, so we need to forget the level of carelessness we're used to when dealing with consumer electronics,” says Mather. .
One way to solve this problem, says Zhang, is to have extra budget and duplicate components on hand. The solution is to have enough money to buy all the duplicates, enough time to wait for parts to be shipped, and enough to try to debug the failure if there is no obvious reason why something broke. Just have patience.
Finding time to work on the project was also a hurdle, Zhang says. Both Mather and ChatGPT helped alleviate this problem, and “having ChatGPT help with the coding was really important,” she says.
ChatGPT can also provide a shoulder to lean on when a project gets tough, Mather says, by helping you debug your code and being a place to vent.
The best tool or app for your AI project
Beyond ChatGPT, Zhang recommends playing around with other accessible and open source models from OpenAI and Replicate to get a feel for how the project works before getting into the nitty-gritty. Masu. She also recommends Replit for prototyping software without a full development setup.
In addition to Teachable Machine, Khalipina recommends using GitHub Copilot to aid in the coding process. And most recently, she says, Vittascience AI is helping her add markers of disease, such as cough.
“We like the idea that a link to a training model can be used by anyone, anytime, anywhere, without having to buy and install expensive hardware,” says Khalipina. “All you need is a computer and the Internet to be alerted to your fatigue level.”
To stay inspired for new projects, Aguilar recommends keeping an eye on sites like HackerNews and HackerNews. X See what kind of projects people are releasing.
Sarah is a Boston-based science and technology journalist interested in how innovation and research intersect with our daily lives. She has written for numerous national publications, and innovation her news is featured on the following sites: reciprocal.