[machine learning generated headline] | August 18

How Not to Insight 

The slow-moving dumpster fire that is the Warner Bros Discovery merger has been the shiny object of internet derision for a few weeks now, but the CW is also a hot mess and should not be overlooked. Like HBO, the CW’s new corporate overlords seem either not to understand scripted entertainment or actively hate it. They also seem to have no clue what audiences want or who their audience even is. 

From Deadline: 

“The demographic focus of the CW will also change over time, Carter said. Historically, shows like Riverdale, All American, Arrow, and Supernatural have focused on viewers in their teens through their 30s. The reality, though, is that the average CW viewer is 58 years old, and Carter said that schism explains why the CW is the lowest-rated broadcast network.” 

So either 58 year olds are the ones actually watching the teen-focused content, or a bunch of teens have their parents’ Comcast logins. Either way, “cancel all the teen dramas” doesn’t feel like the best way to dig yourself out of that lowest-rated spot. 

Platform Updates

Instagram 

• Reels Scheduling is Now Available to All Users in Meta’s Creator Studio App 📅

• Meta Launches New Reels Features, Including Stories to Reels Conversion and Improved Analytics

The Rest of Meta 

• Facebook begins testing default end-to-end encryption on Messenger 🔐

• Anyone can now cross-post Reels from Instagram to Facebook

TikTok

• TikTok now offers a very basic text-to-image AI generator directly in the app (more on text-to-image AI below) 

• TikTok Launches New Shopping Ads as it Continues to Push Ahead with its eCommerce Integrations 🛒

Twitter

• Twitter Outlines New Ad Improvements in Line with Evolving Data Privacy Approaches

YouTube

• YouTube Will Now Automatically Add Watermarks to Downloaded Shorts Clips

LinkedIn 

• LinkedIn rolls out new tools to give creators more ways to share visual content

Pinterest 

• Pinterest Launches New Streamlined In-App Purchase Process for Shopify Merchants 🛒

Snap 

• Snapchat already has 1 million people paying for Plus

• Snap is giving up on its Pixy drone after just four months (grab one while you can, it’s a fun toy) 

Reddit 

• Reddit Announces New Developer Platform to Facilitate Additional Tools and Functions

Tumblr 

• Tumblr Launches “Tumblr TV,” A GIF Search Engine With A Full-Screen Viewing Mode

Culture Movers 

Film & TV 

• She-Hulk did a clever activation on Tinder 

• Get your streaming subscriptions here: YouTube is reportedly building a “channel store”

• HBO Max to Remove 36 Titles, Including 20 Originals, From Streaming

• Streaming Viewership Beats Cable For the First Time

Creator Economy 

• Want your brand to stand out on social? Hire a creator as your Chief TikTok Officer.

• Twitch streamers traumatized after four ‘swattings’ in a week

AI Art Is Here

AI generated art has arrived, and as with any new technology, so has the conversation about how to ethically deploy it. For those of you new to the concept, here’s a great primer on how the machine learning technology behind these computer generated images works. In practice you’ve probably seen images like these that were built with a free version of the Dall-E software. 

There’s a whole Twitter account devoted to these delightfully surreal images after “how weird can I make my prompt for that AI art thing” became a meme a few months ago. While there’s something hilarious about the blurry-faced dream-like images of the free versions, the pro tools can be much more convincing. 

Now many human artists are rightfully concerned. There are two central critiques:

1. Will AI-generated images replace some of the paid work available for artists and illustrators? 

2. How are these machine learning programs being trained? Are artists being compensated when their work is used as training data? 

Recently, writer Charlie Warzel became the focus of these questions after he used an AI-generated image for his newsletter. Warzel published a great follow-up about his own thinking on the subject, why he will be avoiding AI art for future newsletters, and the dynamics of being a Twitter main character. 

“I don’t think I’ll be using Midjourney or any similar tool to illustrate my newsletter going forward (an exception would be if I were writing about the technology at a later date and wanted to show examples). Even though the job wouldn’t go to a different, deserving, human artist, I think the optics are shitty, and I do worry about having any role in helping to set any kind of precedent in this direction. Like others, I also have questions about the corpus used to train these art tools and the possibility that they are using a great deal of art from both big-name and lesser-known artists without any compensation or disclosure to those artists.”

There’s another side of this conversation that’s more hopeful about the possibility of AI art becoming a tool that enhances the work and workflows of human artists. Machine learning tools could be deployed to simplify, scale, and inspire real human creativity. While this optimism does feel out of touch in a post-Facebook tech industry (the man who published this image does not value artists), I am curious to see how marketers will adopt this technology. Could creative teams use AI tools to get approval on concepts before developing new work? Will copywriters and designers send each other AI images back and forth while brainstorming around a brief? Will Meta add a text-to-image generator to its ad platform? Will we all learn to spot subtle signs of machine creative mimicry? Will boldly human work stand out? 

Lots of questions that will make you feel like Don Draper trapped in Blade Runner. 

Update Your Product Marketing Strategy 

The bar has been raised. There is a new gold standard. This is how all patch notes should work going forward.