Gen AI meets art

Inspiration

On my journey as a machine learning engineer and using AI for good, I decided I didn't need another course, I'll build my own projects and start the job before I have it and by doing this I'll make machine learning jobs come to me, so I started building a cutting-edge startup specializing in building AI, Cloud, and Mobile applications for governments, startups, and businesses in Africa. Our mission, to empower organizations to leverage advanced technologies and propel their growth while aligning with the United Nations' Sustainable Development Goals 3, 4, and 9 and I just released a mobile application for a client and it's on play store actually.

What it does

As machine learning finds increasing use in virtually every industry, there’s a growing need for people to bring machine learning models into production, enter Vertex AI, an end-to-end development platform to build, deploy and manage machine learning models, understanding this production cycle is important because it gives you an idea of what work needs to be done to bring a model to the real world.

Vertex AI offers a model garden that includes foundation models. The language foundation models include PaLM API for chat and text. The vision foundation models includes stable diffusion,which has been shown to be effective at generating high quality images from text descriptions.

How we built it

In the past few months, it has basically been all about using the OpenAI API and bringing that into applications so I have got all of the power of the openAI right there in the application, and there are actually a million ways that I could use that in the app! You have probably seen millions of APIs, but I cant think of one as insanely powerful as the OpenAI API, it just gives so much potential and there is so much we can do with it. The OpenAI API include the chatGPT models for language generation, codex for code generation and DALL_E for image generation powered by Gen AI. All I really what to show in this essay is how powerful this generative artificial intelligence is.

Challenges we ran into

Our prompt design process was tideous When using chatGPT as a coding assistant it is important to pass the right content, remember the way we said Gen AI models are trained on various kinds of data, so its results can be quite generic. However by providing more specific information, we can achieve better results, this is the key

Accomplishments that we're proud of

One thing I want show is how all this power of Gen AI is available to people who do not necessary need to know all that much code. You do need some programming skills but it is not terribly complicated programming, if you been studying programming for a couple of months, you can do an awful lot with some of these Gen AI APIs.

What we learned

It is a not a wise decision to rely on code generation like chatGPT to detect security problems within your code. Like when I asked chatGPT to setup our nodejs back-end in our full-stack app development use case, it generated code with a security loophole. Always look at code generation as a suggestion not as gospel truth

What's next for Gen AI meets art

Teach and share the general public on how to be responsible by design and develop these technologies with ethics in mind because with great power comes great responsibility.

Built With

• bard
• chatgpt
• palm