🚀 Is AI art ethical?

Plus, the next phase of the Elon Musk / Twitter acquisition!

Justine & Olivia Moore

Oct 9

▷ Listen

SAVE

trending 📈

💰 Twitter acquisition moves forward. After a several month delay, Elon Musk is moving forward with his bid to buy Twitter at the original price of $44B. Musk filed documents with the SEC, stating that he would move ahead with the acquisition. He also asked Twitter to end its lawsuit against him (scheduled to go to court on Oct 17). The company did not agree to this, but a judge ruled that Musk can avoid a trial if the acquisition closes by October 28. Musk tweeted that he believes buying Twitter will be a three-to-five year accelerant for “X, the everything app” that he plans to build.

🛍️ Poshmark gets acquired by Naver. P2P shopping marketplace Poshmark, which IPOed in 2021, will be acquired by South Korean search company Naver for $1.2B in cash. This implies a $17.90 share price - a 15% premium to where the stock was trading prior to the announcement, but down more than 50% from the IPO price of $42. The acquisition appears to be part of Naver’s efforts to expand internationally, as the company purchased its first North American business, Wattpad, in 2021.

Check out Olivia’s analysis of the Poshmark vs. Depop acquisitions here! (Note: this was published when the deal price was being reported as $1.6B, not $1.2B).

⚖️ Kardashian hit with crypto fine. The SEC announced this week that it charged Kim Kardashian for “unlawfully touting” a cryptocurrency, in relation to a 2021 Instagram Story post about the Ethereum Max token. Kardashian was paid $250k for the ad, and settled with the SEC for $1M in penalties in addition to the original payment plus interest ($260k), totaling a $1.26M fine. Kardashian did label the post #ad, but the SEC’s Director of Enforcement stated that crypto ads must be labeled with the “nature, source, and amount of compensation” received.

what we’re following 👀

Why on-demand marketplace startups are shifting towards pre-scheduled orders.

Three directors of EV startup Faraday Future have redesigned due to death threats.

Canned water startup Liquid Death is now valued at $700M! (see our past writings on Liquid Death here).

If you’re looking to break into VC, check out this Contrary Capital webinar on 10/18.

gibsly🔪👻 @slopmiser

ai art was fun when it was just a bunch of people putting shit like "jeffery epstein funko pop mayan tomb" into a big machine for a laugh but now its just like 7 guys who circle jerk eachother for being "real artists" because they can type "astronaut lady beatiful lighting"

Suhail @Suhail

There's no doubt in my mind that making AI art is real art. I spent about 1.5 hours tweaking things to produce these visuals. The hair, perfect red lipstick color, focus, eyes, wrinkles, theme, reflections, clothes. It was a joy to achieve a result I couldn't have previously. https://t.co/w4vwvhtiYa

It’s been just over a week since the public release of generative AI platform DALL-E, and the debate over AI-produced art is reaching a fever pitch.

For the uninitiated: DALL-E generates images based on natural language descriptions. You can type something like “Labradoodle wearing a wizard hat and cape in the form of an Impressionist oil painting,” and get outputs like below. After an image has been generated, you can then edit it in DALL-E itself or in an external app.

While DALL-E may be the most well-known (for now), it isn’t the only generative AI art tool out there. If you follow App Store data, you’ve probably noticed a surge in downloads for products like DREAM by Wombo, Starry AI, and Wonder (parent company Codeway) - all of which take text and turn it into art. They join web-based platforms like Midjourney, Nightcafe, and Hotpot AI. Last month, an artist won the Colorado State Fair’s digital art competition using a Midjourney-generated piece.

Why is this generating so much controversy? After countless hours browsing Twitter and Reddit, we’ve identified two main arguments used by opponents:

AI art is “soulless.” The creator provides guidance around the content and style of the image, but isn’t actually creating it. Proponents of this argument often say that what imbues meaning in art is the countless little decisions made by a human in the process of designing a piece - their emotions and motivation are critical. DALL-E doesn’t feel anything when making the image, so it can’t be art.
AI art steals from “real” artists. How do you teach AI to create art? You train it on millions of images of the real thing. DALL-E allows users to specify if they want to mimic the style of a specific artist in the image prompt - Greg Rutkowski is one popular choice. This has raised questions about the line between inspiration and theft, especially given that artists don’t benefit from the use of their work in AI image generators (but the creator can sell the resulting image!).

Fans of AI art would likely counter this by saying that the creator is making choices about the content and style of the image. They might have conceptualized a beautiful piece of art, but aren’t able to create it because they lack the technical skill. In this way, AI is democratizing access to art, and the result will be more art in the world and more people who can feel fulfilled as artists.

On the topic of “theft,” a counter-argument is that almost everything created today is somehow inspired by something else. Even in the real world, artists play off of or imitate each others’ work. This is how we end up with movements like Impressionism and Pop Art - they aren’t driven by one artist alone, but by a group of people who like the same style or subjects. To take this to an extreme, Pablo Picasso once reportedly said: “good artists copy, great artists steal.”

If we assume AI art continues to gain popularity, this opens up a lot of interesting questions - here’s a few we have:

How will the rise of AI impact different types of artists? High-end art, where the value comes from the identity of the artist and the provenance of the work as well as the piece itself, may be relatively insulated from any AI “threat”. More commercial artists and designers might be more heavily impacted.
Will any AI artists rise to fame? In the traditional art world, artists are able to “break out” even if their style was eventually mimic-able. However, it seems like a world where anyone, regardless of skill, can instantly create a duplicate is different. Will AI artists be able to build brands? (We can imagine a hilarious guarding of the “keywords” someone used to create a certain design 😂)
Can AI art have a provenance? How will we track the origin and creator(s) of each piece of AI art? Is it whoever publishes it on the Internet first? This is particularly tricky as the same piece could be generated across different AI art platforms, so there’s no one platform that serves as a “system of record.”
- Perhaps crypto is the answer here, with individual pieces of art published as NFTs. In that case, even if someone creates an identical piece, there’s a record of who originally generated it!