5 min read

2.7 Trillion AI-Generated Images Will Exist by 2027

We’re flooding the Internet with AI-generated content at an unbelievable rate and quickly approaching a point where the Internet has more stuff made by AI than humans.

We’re not just talking about art from Stable Diffusion, DALL-E, and Midjourney. Literally, any type of media has a generative AI out there which has already scraped millions of data points in that vertical and learned how to generate new data from it.

That’s not even the half of it. Futurepedia documents hundreds of AI tools for generating any type of content you need. The result is that we’re not just augmenting our visual media creation but just about everything else that exists on the Internet.

Many people won’t just enhance their abilities, they’ll replace their presence — using the tools at every chance. If we can use these tools for any creative activity, many (if not most) will use them for all creative activities. – Alberto Romero

About a month ago, Suneet Agarwal went viral on Instagram for sharing his process for creating content using AI:

  1. Finds a trending non-fiction topic on YouTube
  2. Runs the video through Otter.ai to generate a transcript
  3. Runs the transcript through Jasper.ai to create a new video script for himself
  4. Films the video and then runs the script back through Jasper.ai to create a net-new blog post
  5. Runs that blog post back through Jasper.ai to make multiple short-form posts

Yes, it’s unoriginal. And his comments really let him have it (I’ll get to that later). But this shows how easily AI will flood the Internet with content.

The Volume of AI-Generated Content

In Alberto Romero’s blog titled, “Generative AI Could Pollute the Internet to Death,” he outlines the current output of AI-generated images and plots the growth of this industry to predict how long it will take for content generated by AI to overtake content made by humans.

(TL;DR – In 5 years, AI will have generated 4x more images on the Internet than humans have in the entire existence of the Internet or about 2.7 trillion images against 750 billion. And all other forms of content creation will likely follow this path.)

On October 17, TechCrunch reported that 1.5M people had signed up on DreamStudio (the official UI for Stable Diffusion — now one among many). In total, they’ve generated around 170M images. Emad Mostaque said “across all channels” Stable Diffusion has +10M daily users.

If we extrapolate the number of images per user per day from DreamStudio data to the 10M total users (a conservative assumption given that those who use other UIs are likely more deeply involved), we have that people have created around ~1.1B images with Stable Diffusion. In 2.5 months.

Midjourney, another popular text-to-image model, has 3.7M Discord members. It opened the beta in March. If we take DreamStudio’s numbers and assume a similar growth rate (a reasonable assumption given that both subscription models are similar), we have another ~1.3B images.

With OpenAI’s DALL·E (1.5M sign-ups creating 2M images/day since September) we can easily add another 200M (the model was announced in April but OpenAI let in people slowly throughout the months).

In total, that’s 2.6B images created in half a year with three models. There are ~750B images on the internet. If my assumptions are correct, around 0.35% of all images on the internet are now AI-generated.

Not many. Yet.

Now, imagine that Stable Diffusion grows in a linear fashion (probably a conservative assumption) during the next 5 years to 1B daily users (a plausible total growth if we assume the tech will mature and be integrated into popular products and services).

Under those conditions, assuming everything else is constant and accepting my previous assumptions, we’d have ~2.7T AI-generated images by mid-2027.

That’s four times all human-made images on the Internet. And a similar future can be expected of text, code, etc. – Alberto Romero

Is This Good or Bad?

Many argue there’s no such thing as original thought and whether from AI or humans, there is always input copied from elsewhere. Furthermore, with visual media, AI already touches a lot of what we produce, whether we’re talking about filters, face-tune, or some of the other tools offered in Photoshop. Still, there’s inherently something different between a filtered photo and a DALL-E-generated image.

The issue is that AI enhances the volume of what we produce and publish on the web. What’s the result of a world where AI-generated content greatly outweighs human-generated content?

  1. Is it just an inconvenience, like an email inbox with too much spam and marketing junk?
  2. Does it stand to change our beliefs and interpersonal connection to each other, like a social media feed overtaken by bot-generated, divisive language?
  3. Or is it actually an improvement, like a Renaissance of human creativity assisted by AI?

Ideally, scenario three is where we end up (and a select few with great intentions will spark a Renaissance). But we’re mostly going to experience the first two scenarios.

The Internet already contains more info than any human (or all combined, for that matter) could ever consume in a lifetime. Just on YouTube, users upload 30 years’ worth of videos every day.

It’s hard to argue we need more. Still, we have the incentive to create more data because the majority isn’t intended to convey or store ideas, thoughts, or feelings, but to attract attention (to achieve some further goal).

AI-generated data, due to sheer quantity, could erase the very purpose of the Internet—keeping large amounts of high-quality info at the surface, easily accessible. AI could bury high-quality, intent-driven human-made creations under tons of generic and bland content. – Alberto Romero

The problem is that generative AI will mostly produce low-quality data (junk we don’t care about), do so at an unfathomable rate, and result in irreversible Internet pollution.

Take a platform like DeviantArt, which is one of the largest image hosting and portfolio sites for digital artists. When they announced the launch of DreamUp, a tool for text-to-image generation, their community was up in arms. And for good reason too. The people whose work has unfairly been used to train generative AI are now faced with competition from an AI tool natively on their portfolio site. Inundating that creative community with AI-generated content seems like a travesty.

Look no further than the artist Greg Rutkowski, whose name has been used as a generative image prompt at least 93,000 times to date.

“It’s been just a month. What about in a year? I probably won’t be able to find my work out there because [the internet] will be flooded with AI art,” Rutkowski says. “That’s concerning.” – Everydays

AI-generated content is flooding the Internet. Whether or not this flood will wash away everything we’ve built is yet to be seen.