AI Art in Film and Music: SN13 Sentiment Analysis

Artificial intelligence has entered the entertainment industry - but how do people really feel about it? An SN13 dataset analysis report.

AI is everywhere. It’s in our worklives, across our social media, causing questions of legitimacy in news reporting, and in academic fields like mathematics and physics. Some of these affect us directly, others are far removed. But for most, AI exists in a specific place that seems to intrigue or captivate practically every human being: art.

AI image generation was one of the first consumer-facing uses of the modern era. A relatively early example of this is Google’s DeepDream, an image creation system that became known for building works that had a dog-like pattern to them. A later example that left people awe-struck was the 2019 website thispersondoesnotexist.com, which used AI to generate unique faces on every refresh.

At the time, those models fostered a near-universal level of intrigue. We wanted to know what would come of the industry. Since the release of OpenAI’s DALL-E in 2021, the perspective has changed. At that point, AI models were getting startlingly impressive, and at a rapid rate. While everybody could tell what was genuine and what was fabricated back then, it only took a few more years for people to question the legitimacy of what they were seeing. In 2026, AI is able to build highly convincing creations - these can look like real-world depictions, or they can look like art that’s been made by human hands and minds.

It’s not just images. AI is getting strikingly better at video and music generation. The arts have been effectively infiltrated by these machines. The creative industries are opening up to them. However, what do people truly think of this? Is the excitement from years ago still there? Or are there concerns about the authenticity of the future?

Subnet 13, Data Universe, has been monitoring the discourse over AI in film and music for the last three months, using our real-time data collection subnet to perform social listening on X and Reddit. We have two datasets designed specifically for this, which are available on our marketplace to purchase. See our AI music dataset here, and our AI film set here.

We’ve downloaded the data, and performed our own sentiment analysis on the topics, to give you a direct signal on how people think, feel, and communicate their stance.

How to Extract Sentiment From Data Universe?

SN13 provides you with the raw data. To make use of it, simply purchase and download the CSV files from the datasets, and ask a high-grade consumer-level LLM for insights, including graphs and charts. You can specify timeframe, the types of charts you’re looking for, and even ask your model of choice to highlight specific quotes or posts which it deems as pertinent to the conversation.

For large datasets, we recommend using Claude, as it has a higher context window.

Emotion Trends for AI Music and Film: Overview

The following charts show a 7-day rolling average for emotion trends regarding AI music and film. Both charts span the same time frame (October 17 2025 to January 21 2026). They also cover sentiment from X and Reddit based on five key markers: excitement, fear/anxiety, anger/outrage, sadness, and surprise.

Throughout their time, excitement appears to be the dominant trend. From digging through the data, we see many posts from people who feel inspired or forward-thinking about the industry. Let’s take a look at some examples.

“This is one of the best seven minutes that I spent on AI film. Wonderful storyline, perfect visuals. I loved the Suno track” - in reference to this video.

“Just got back from the Arizona AI Film Festival feeling fully energized, deeply grateful, and inspired by beautiful new friendships and an elevated vision for the future. It was truly uplifting to meet like-minded, passionate people and to be surrounded by such creative energy” - in reference to this event.

“Here are some great AI artist accounts that we enjoyed interacting with today: @Stylez_Morales (music and film) @muriellondon (fantastic Iranian woman artist)... We especially enjoyed being a guest of @rebirthofthewo1 on his real-life podcast”

“Inspired by some of my favorite chillout music from the early 2000’s. Some nu jazz, trip hop and electronic. Bringing chill back. #AIMusic #SunoAI #AIart #aiartcommunity #aiartist”

Tip: To source specific posts from your raw data, simply ask your LLM of choice what types of posts you’re looking for. You can cross-reference to ensure accuracy by searching for the exact quotes within your CSV files.

—-

However, there are two moments worth dialling into.

AI film: January 2026 Decline in Excitement

During mid-January 2026, excitement in AI film took a nosedive, with surprise reactions leaping ahead.

Around this time there’s an inundation of negative posts:

“Lol ‘AI filmmaker’ that’s a straight up oxymoron”.

“There will be a fully AI movie submitted to indie film competitions as early as next year... Actors will be able to make money by selling their likeness to movies while literally everyone else loses their job”.

These are expressly anti-excitement sentiments, which emerged at the time. Alongside this, the actor and director Natasha Lyonne faced a wave of criticism in this period for her support of AI in film, and the fact she owns her own studio in the field. Posts denigrating her position include:

“I was a big fan of you I really was but you know it kinda all went down the drain when you started a ‘ethical’ Ai movie production company #FUCKAI”

“Don’t try to be cute, you tried making an AI movie and then claimed David Lynch would have supported that – dude wasn’t even cold!”

“Imagine siding with a bunch of silicon valley ghouls over actual people”.

There doesn’t appear to be any particular reason why she attracted so much attention in January, other than the fact she’s well-known, and the X algorithm may have distributed her words far and wide.

That being said, the rise in negative sentiment could be attributed to the Google AI Film Awards, which announced their winner, Zoubeir ElJlassi, on January 14th 2026. This seems to have triggered some critical responses, questioning what it means to be a filmmaker in 2026, and whether the role of AI will usurp human artists.

While excitement faded, surprise rose. In our report, surprise is shown as a sign of shock or awe, not inherently positive or negative, but more so mind-blowing or disconcerting.

At that period we saw the following posts:

“AI film looks totally wild now”

“Crazy We’ll have a full AI movie”

“AI film making is getting to an insane level. Enjoy 😊”

“Wont be long before we can all make an ai movie. Thats gonna be wild”

“Crazy what is happening with AI. Was having a talk with my Dad about how AI is developing like crazy and all these ads popping up saying they were taking over UGC & my Dad made a good point. The first AI movie probably isn’t that far away either”.

One factor these posts have in common is that, while they initially appear to express excitement, on closer inspection they speak more about the impressiveness of AI’s capabilities, rather than any inherent positivity with the medium. They’re makers of competence, not appreciation. A simple sentiment analysis chart mapping only positive and negative emotion may have missed that.

Those may have increased due to the recent $130 million raise Higgsfield AI raised, announced in January, which could have caused users to think deeply about the social impacts. Google also announced its latest Veo update, its video-creation model, which provided greater control and consistency. Together they could have led people to contemplate the industry’s future.

AI music: December 2025 Excitement Turns to Sadness

AI music has its own perplexing activity, this time in December 2025. Throughout the three months we captured, excitement dominated, however, by the end of 2025, it was briefly matched and overtaken by an uptick in sadness.

Posts marked as sad are ones with negative emotion, often critiquing the state of the affairs without particular anger or disdain, but rather a depressive feel. In that period, the following posts were made:

“Could you imagine ten years from now there’s a Grammy category for best AI artist? The rock/metal category winners would probably still be handed out their awards in the separate, bland auditorium with empty seats. I’m hearing entire full length AI movies are emerging too. It sucks”.

“Had to sit there, look at an AI image and listen to AI music, because my brother would rather go on youtube and search up ‘christmas music radio’ 😱😱😱 at the family function. Art truly is dying huh”

“Artists don’t get paid much for albums... The music industry is surely already considering the idea of propping up AI artists to replace human ones”

Those posts cover both a worry around the nature of the music industry from an artist’s perspective, and disdain for engaging with AI music rather than human-produced work for the listener.

During that time, UMG, one of the world’s largest music corporations and record labels, announced a collaboration with Spice, an AI music creation company. Alongside this, The Guardian published an examination of the current AI music landscape, questioning why labels were embracing it whilst artists feared it. This could naturally have led to more people feeling dejected about the future ahead.

Post frequency metrics

Let’s look at the frequency of posts within our timeframe.

The following chart shows the daily post volume within our datasets. Here, you can see a steady rise in the count over time, with a peak in early January 2026. This covers both AI film and AI music.

For a post count on AI film, this is what our data reveals.

For a post count on AI music, the data is here.

Tip: To collect daily volume metrics from our datasets, ask your LLM of choice to plot the count on a chart, using the post count as its Y axis, and the data as its X axis.

Joy and Disdain Within AI Art

Our data reveals two important factors: people are in awe of what AI’s capable of in the current era, but they’re fearful of where human involvement fits in. AI has inspired those who use it, and empowered them to create impressive works without conventional technical capabilities, but in the same sense, it’s caused those who create art organically to feel existential discomfort about their future.

There’s a tension between companies and studios wanting to unlock new innovations (often for cost-saving measures), and individuals who worry their livelihoods will be eroded. It hits a greater question around the role of art in life, and whether AI’s presence in there impedes on our ability to create, appreciate, and earn a living in the process.

All of this was gathered from Data Universe’s AI film and AI music datasets, available for purchase on our marketplace, along with an LLM for gleaning insights. Download them yourself to give it a try. Or browse through our many other real-time updating X and Reddit datasets for sale, all touching in pressing real-world problems in the current day. These are perfect for marketing, social listening, and sentiment analysis.