“Generative AI” is booming, and it’s revolutionizing the way people make art — and money

Memes become Monet paintings… A new kind of AI is transforming the way people create art and do business. Generative AI uses deep learning and tons of data to convert text prompts into images, videos, and articles. We made the image above using gen-AI and a text prompt.

• Main character: OpenAI, which has raised billions from Microsoft and others, launched the first major gen-AI tool, Dall-E, last year. It was released to the public last month, and Reddit’s already filled with viral Dall-E creations.
• Other players: Stability AI raised $100M this month (at a $1B+ valuation) for its open-source gen-AI tool, Stable Diffusion. Meta, Google, Salesforce, and Adobe have built in-house gen-AIs, too, but they’re not publicly available.

Goofy images were only the start… Now entrepreneurs are putting gen-AI to work. 140+ gen-AI startups have launched in recent months: Jasper, which makes AI-generated blog posts, raised $125M at a $1.5B valuation last week. Companies can use it to automate copywriting (think: “a promotional post about air fryers in a friendly tone”). Other startups are auto-generating clothes, floorplans, and ad jingles. Microsoft added Dall-E to its design tools, and Shutterstock plans to use it to make stock images. But ethical concerns abound:

• Payment problems: Critics say gen-AI doesn’t compensate artists who create the content that “feeds” its algorithms (though Shutterstock has a “contributor” fund).
• Harassment hitches: Dall-E has safeguards, but some argue that gen-AI can still be used to produce celeb deepfakes and misinfo. Others say it replicates biases, like sexist imagery.
• Automation anxiety: Some fear gen-AI will put writers and designers out of work (others disagree).