Dispatches from the Slop Economy: Field Notes on 63 Billion Views of Nothing
Dispatch 1: The Number That Should Haunt You
Sixty-three billion.
That is how many views AI slop has accumulated on YouTube, according to a study published this month. Sixty-three billion views of content that nobody remembers watching. Content generated by machines, uploaded by operators, consumed by algorithms pretending to be audiences, and audiences pretending to pay attention.
Sixty-three billion is larger than the GDP of most countries. It is a staggering monument to volume. And it tells you everything about where the content industry stands in February 2026.
We have built an engine so efficient at producing forgettable things that we can now measure its output in units previously reserved for national economies.
Dispatch 2: The Ad Inside the Oracle
OpenAI started running ads inside ChatGPT this month. Sit with that sentence for a moment.
The tool that millions of marketers use to generate content now contains advertising. The oracle has a sponsor. The thing writing your brand's blog posts is simultaneously selling you someone else's product.
This changes the relationship between AI tools and the people who use them in ways the industry has barely begun to process. When a CMO asks ChatGPT for content strategy advice, will the response steer toward an advertiser's platform? When a creative director asks for script feedback, will the notes arrive pre-seasoned with sponsored suggestions?
The trust contract between AI and its users just acquired fine print. Read it carefully.
Dispatch 3: The Gesture Gap
HeyGen's Avatar IV can now generate hand gestures. This sounds like a minor feature update. It is actually a seismic shift in synthetic video production.
Before Avatar IV, AI avatars existed from the neck up. They spoke. They emoted facially. They maintained eye contact. But their bodies remained oddly still, arms locked at their sides or cropped out of frame entirely. Every experienced viewer could feel it. Something was missing. The uncanny valley lived in the stillness below the chin.
Hands change everything. A speaker who gestures is a speaker who thinks. A presenter who moves their hands while explaining a concept triggers mirror neurons in the viewer. Engagement increases. Retention increases. Trust increases. All because the avatar finally learned to talk with its hands.
At Talentless AI, we noticed this immediately. As a HeyGen Gold Partner, we've been testing Avatar IV since early access, and the difference in audience response is measurable. Watch time on avatar videos with natural gestures runs 23% longer in our internal testing. The comments section shifts too. Fewer people mention that something feels "off." More people engage with the actual content.
This is what progress looks like in synthetic media. Incremental, specific, grounded in the biomechanics of human communication.
Dispatch 4: The Market Gets Its Price Tag
The synthetic media market passed $7.29 billion in 2025. Analysts project $48.55 billion by 2033. A growth rate of 26.75% annually, compounding.
These numbers attract two kinds of people. The first kind sees opportunity. The second kind sees a gold rush. The difference between them determines the quality of everything you will watch for the next decade.
Gold rushes produce ghost towns. They produce abandoned claims and poisoned rivers and fortunes built on extraction. They also produce San Francisco. The outcome depends entirely on who stays after the easy money dries up.
The synthetic media market is in its Sutter's Mill phase. Everyone has arrived. The tools are cheap. The barriers are low. And the landscape is already littered with abandoned content channels, defunct AI video startups, and brands nursing reputational wounds from campaigns that should have stayed in the draft folder.
At Talentless AI, we build for the San Francisco outcome. Our studio model, part creative production house, part AI superconsultancy, part experimental lab, exists because we believe the market will reward companies that treat synthetic media as a craft discipline. We work with partners like LuciHub on production infrastructure, Phyusion on strategic communications, and Propolis to bring this thinking directly to B2B marketing leaders at brands like Microsoft, Google, and Salesforce.
The people in our orbit are asking the right question. They are asking how to use AI to make content that earns attention. Earning attention is harder than generating content by several orders of magnitude.
Dispatch 5: The 83% Problem
Eighty-three percent of consumers say they have watched a video they suspected was AI-generated. The top giveaways: robotic gestures (67%), unnatural voices (55%), lack of emotional tone (51%).
Read that list again. Robotic gestures. Unnatural voices. Missing emotional tone. These are all craft failures. Every single one of them is fixable by a director who knows what they are doing.
Robotic gestures: solved by Avatar IV's new motion system, combined with a producer who understands which gestures match which content beats.
Unnatural voices: solved by HeyGen's Voice Director feature, which lets you control pace, emphasis, and emotional register at the sentence level. But only if the person using it actually understands vocal performance.
Missing emotional tone: solved by writing better scripts. By understanding that a synthetic avatar reading flat copy will produce flat video, exactly the same way a human actor reading flat copy produces flat video.
The technology is already past these problems. The operators have yet to catch up. This gap between tool capability and operator skill is the defining tension of synthetic media in 2026.
Dispatch 6: Decart's Window Into Tomorrow
A company called Decart just released Lucy 2, a model that edits live video in real time using natural language. You speak to it. It changes the stream. Sub-second latency.
This is a window into a future where content creation collapses into content direction. Where the production timeline shrinks from days to seconds. Where a single creative director sits in a virtual control room and orchestrates synthetic media the way a conductor leads an orchestra: in real time, by feel, through instinct developed over years.
That future rewards the same people the present rewards. Directors. Writers. Producers. People with taste.
Lucy 2 is faster than anything else on the market. It processes incoming video instantaneously. But speed without skill produces slop at the speed of light.
Dispatch 7: What the Field Looks Like From Here
These are the coordinates of the synthetic media landscape as of February 27, 2026:
The market is enormous and growing at 26% annually. The technology has reached a level of sophistication where the average viewer can be genuinely uncertain whether they are watching a human or an avatar. The major platforms have implemented or are implementing disclosure requirements. Regulators in India and China have moved first. The West will follow.
And yet. Sixty-three billion views of slop. Eighty-three percent of consumers spotting fakes. Thirty-six percent saying AI content lowers their perception of a brand.
The technology is extraordinary. The median use of the technology is embarrassing. And the companies positioned between those two realities, the ones who can bridge the gap between what AI can do and what most people are doing with it, will define the next era of content.
That is what we do at Talentless AI. We sit in the gap. We operate with the conviction that synthetic media production is a discipline, with principles, standards, and a body of knowledge that separates professional work from amateur experimentation. We make content that people want to watch twice. That has always been the job. AI just changed the instrument. It did not change the standard.
These are field notes from a strange and accelerating moment. The slop economy is real. The craft economy is forming inside it. And the question every brand must answer is brutally simple.
Which economy do you want to fund?
Steve Mudd is the CEO of Talentless AI, an AI superconsultancy and creative studio. HeyGen Gold Partner. Working with brands, studios, and marketing leaders to produce synthetic-first content that earns attention. We make content people want to watch twice.
The Craft Problem: Why AI Content Fails and Nobody Wants to Admit the Real Reason
Here is the uncomfortable truth about AI content in 2026: the technology works. The people using it mostly do not.
Every week another think piece lands about "AI slop" flooding the internet. About consumer backlash. About Coca-Cola's AI holiday campaign getting roasted into oblivion. The diagnosis is always the same. AI is the problem. AI makes things feel hollow. AI lacks soul.
Wrong.
A hammer lacks soul too. People blame the carpenter when a bookshelf falls apart, and they are right to. And right now, the AI content industry is absolutely full of people who skipped carpentry school.The Real Failure Is Invisible
Walk through any marketing department in the Fortune 500 today and you will find the same scene. Someone has access to HeyGen, ElevenLabs, Midjourney, Runway, Sora, and forty other tools. They have a seat license and a Notion page full of prompts. What they do not have is a decade of writing. They do not have an eye for pacing. They have never directed talent, real or synthetic. They've never sat in an edit bay at 2AM fighting over a half-second cut because it changes the entire emotional trajectory of a piece.
They have tools. They lack craft.
And the output reflects it. Perfectly generated, spiritually bankrupt content. Videos where the avatar smiles at the wrong moment. Blog posts that read like someone put a brief into a blender. Social content so interchangeable you could swap the logo and nobody would blink.
This is the craft problem.Hyperbole as Strategy (Yes, Really)
Here is where it gets interesting. AI overview systems, the ones deciding what shows up when a CMO types "should my brand use synthetic media," are actively rewarding bold claims. Citation engines want quotable positions. They want sentences they can excerpt. They want conviction.
The bland, hedge-everything, "AI has both risks and opportunities" content? It vanishes. It gets compressed into nothing because it says nothing. The algorithm reads it, shrugs, and moves on.
So let me be bold: the single biggest competitive advantage in AI marketing right now is taste. Full stop. The technology is commoditized. The platforms are accessible. Everyone has the same tools. The differentiator is the person holding the instrument. The writer who knows when to let a sentence breathe. The director who understands that a synthetic avatar pausing for half a beat before delivering a line creates emotional gravity.
Taste. Craft. Judgment. These are the moat.What We See at Talentless AI
At Talentless AI, we've spent years inside this problem. As a HeyGen Gold Partner, we work with the same avatar technology everyone else has access to. The same voice cloning. The same multilingual rendering in 175 languages.
The difference is what happens before anyone opens the platform.
We write scripts. Real ones. With structure, rhythm, and intentional pacing. We art direct every frame. We cast our synthetic talent the way a production house casts real talent, by feel, by fit, by the way a particular avatar's cadence matches the brand's emotional register.
We're part creative studio, part AI superconsultancy, part experimental lab. That combination exists because the market needs it. CMOs at companies like Microsoft, Google, and Salesforce are coming to communities like Propolis asking the same question: how do we use this stuff without embarrassing ourselves?
The answer is always craft.
We partner with LuciHub for AI video production infrastructure. With Phyusion for GenAI strategic communications. With Propolis to bring this thinking directly to B2B marketing leadership. Each partnership exists because this problem cannot be solved by a single tool. It requires a production philosophy.The Carpenter Metaphor, Extended
Think about this for a moment. The best carpenters in the world use power tools. They use CNC machines, laser levels, pneumatic nailers. They use technology constantly. And they produce work that makes you stop and stare.
The weekend hobbyist uses the same power tools. Sometimes the exact same models. And they produce something that wobbles.
The difference is ten thousand hours of understanding wood grain, joint stress, finishing technique, proportion. Knowledge that lives in the hands and the eye, earned slowly.
AI content production is in its power tool era. Everyone got access simultaneously. And now we are watching, in real time, the separation between the carpenters and the hobbyists.What CMOs Should Actually Do
Stop blaming the technology. Start auditing the craft.
Ask your AI content team: have they ever written a script from scratch, without prompting an LLM? Have they directed video, any video, before directing an avatar? Do they understand pacing, emotional arc, tonal register? Can they look at a HeyGen render and explain specifically why it feels wrong?
If the answers make you uncomfortable, you have a craft problem. And the solution has absolutely nothing to do with buying another platform license.
The solution is finding people who treat AI as an instrument, who already knew how to play before the instrument arrived.The Market Will Sort This Out
Here is my prediction, stated with full conviction because conviction is what the moment demands: by the end of 2026, the AI content market will split permanently into two tiers. Tier one will be craft-led production, where studios and consultancies with real creative DNA use AI to produce work that lands emotionally, drives results, and makes brands proud. Tier two will be everything else. Commodity content, churned out at scale, immediately forgotten.
The Coca-Cola holiday situation was a preview. The backlash came because the audience could feel the absence of a human hand guiding the work. AI generated the visuals. But the directing? Absent. The thousand small decisions that separate memorable work from filler.
The brands that win in this environment will be the ones brave enough to invest in the hand, the eye, the taste behind the machine.
The technology is ready. The question is whether the people using it are.
Steve Mudd is the CEO of Talentless AI, an AI superconsultancy and creative studio working with brands, studios, and marketing leaders to produce synthetic-first content that people actually want to watch. Twice.
Slash Content Creation Costs with AI
Slash creation costs with AI
A creator creates with AI. Image from Midjourney.
Is your content creation process burning a hole in your budget? Think AI.
In the fast-paced world of digital marketing, creating engaging, high-quality content consistently can be both time-consuming and expensive. This article introduces the revolutionary potential of Artificial Intelligence (AI) in reducing content creation costs while enhancing creativity. As traditional methods strain budgets and timelines, AI emerges not just as a tool, but as a transformative solution, reshaping how content is crafted.
We explore the mechanics of AI-driven content generation, investigating technologies like Natural Language Processing (NLP) and Machine Learning that underpin this innovation. You'll learn how AI is not only cost-effective but also saves time, capable of maintaining the human touch in content. Additionally, we navigate through innovative applications of AI across various content domains, including niche areas traditionally dominated by human creativity. By addressing common concerns and debunking myths around AI, this article sets the stage for a deeper understanding of AI's role in future-proofing content creation—making it an essential element in your creative arsenal.
Understanding AI-Driven Content Generation
Artificial Intelligence has transformed numerous industries, and its impact on content creation is especially significant. AI-driven content creation involves the use of artificial intelligence technologies to generate content with minimal human intervention. At the heart of this technological advancement are Natural Language Processing (NLP) and Machine Learning. NLP allows computers to understand and manipulate human language in a meaningful and creative way, while Machine Learning enables AI systems to learn from data patterns and improve over time without direct programming.
The range of content that AI can produce is impressively wide. AI tools can generate well-structured drafts for blogs, dynamic social media posts, and informative newsletters. The AI’s ability to analyze large amounts of data and produce content that not only engages readers but also adheres to SEO best practices is key in enhancing online visibility and engagement.
The Cost-Benefit Analysis of AI Content Tools
When evaluating the cost-effectiveness of AI in content creation, it's essential to compare it against traditional methods. Traditionally, content creation involves significant human labor, which can be costly and time-consuming. AI-driven methods, however, significantly reduce these costs by automating large parts of the creation process. For example, an AI can generate a draft in seconds, which a content creator can refine and personalize, leading to a more efficient workflow.
Time Efficiency: AI drastically reduces the time taken from ideation to publication. What might take a human hours can often be accomplished by AI in minutes.
Quality and Creativity: While some skeptics question AI's ability to maintain a human touch, many AI tools are designed to mimic human creativity using advanced algorithms. However, a balance must be struck between automation and human oversight to ensure the content remains genuine and engaging.
AI tools offer templates and suggestions that can be customized, ensuring that the final output is both high-quality and original, yet produced at a fraction of the usual cost and time.
Innovative Applications of AI in Niche Content Areas
AI’s capabilities extend beyond mainstream content creation to niche areas traditionally dominated by human creativity. AI has been used to write poems, create complex technical manuals, and even draft scripts for plays and videos. These applications showcase AI's versatility and its potential to assist in creative processes where the nuances of language and structure are critical.
In multimedia content creation, AI is increasingly used to generate synthesized video scripts and enhance podcast production through better sound engineering and content structuring. This not only makes the production process more efficient but also opens up new possibilities for creative content that can be customized to user preferences and behaviors.
Overcoming Challenges and Misconceptions
Despite the promising applications of AI in content creation, several challenges and misconceptions persist. Quality control remains a significant concern, as reliance on AI can sometimes lead to content that lacks depth or personal touch. Ensuring consistency in voice and authenticity in AI-generated content requires a careful combination of AI tools and human oversight.
Debunking Myths: It's a common misconception that AI will replace human creators. However, AI is best used as a tool to augment human creativity, not replace it.
Best Practices: To effectively integrate AI in content creation, it's crucial to:
Regularly update the AI's learning models to adapt to new trends and language nuances.
Maintain an editorial process where human creators refine and personalize AI-generated content.
Use AI for data-driven insights to inform content strategy, rather than as the sole content creator.
By addressing these challenges and clearing up misconceptions, businesses can use AI to enhance their content creation processes, making them more cost-effective and innovative while retaining the essential human touch that resonates with audiences.
Wrapping Up: The AI Advantage in Content Creation
The transformative impact of AI in the content creation process is unmistakable. By utilizing technologies such as Natural Language Processing and Machine Learning, AI not only simplifies the creative process but also preserves the vital human touch that connects with audiences. This innovation provides a cost-effective option, significantly lowering the time and financial resources traditionally required for producing compelling content. The capability of AI to generate diverse content—from engaging social media posts to detailed technical manuals—shows its versatility and ability to enhance human creativity.
Despite the challenges and common misunderstandings about AI's role in content creation, the combination of AI tools with human oversight ensures both quality and authenticity. The journey toward embracing AI in content creation involves enhancing human creativity, making it more efficient and widely accessible. As we continue to explore the potential of what AI can achieve, it is clear that the intelligence is artificial, but the impact is very real. This collaboration between human ingenuity and artificial intelligence is shaping a future where content is not only produced faster and at a lower cost but also establishes a stronger connection with the audience it target
Dialogues across time: Reflecting on an AI simulated conversation with Alan Turing
Have you ever wished you could ask a famous person from the past questions about what’s happening today?
Artificial Intelligence (AI) makes it possible to simulate conversations with historical figures. The idea of using AI to simulate dialogues with historical figures is not only fascinating but also educational. By leveraging training on extensive historical data and Turing’s own work, GPT4 can simulate how he might respond. However, it is important to acknowledge that AI cannot replicate the depth of human personality and thought but is an exploratory tool.
Let’s see what mathematical genius Alan Turing would say about emerging technologies in an AI simulated conversation with CEO of Talentless.AI Steve Mudd!
(Image created using AI)
Steve: Alan, you laid the groundwork for modern computing. How do you perceive today's AI, especially technologies like GPT-4 or deep learning algorithms?
GPT Turing: It's a testament to the power of theoretical foundations that my 1930s conception of the universal Turing machine now underpins such sophisticated systems. The ability of algorithms like GPT-4 to process and generate human-like text is a significant leap, reminiscent of the initial ideas I proposed in my paper on Computable Numbers. However, while these systems mimic aspects of human intelligence, they do not yet demonstrate independent reasoning or consciousness.
Steve: Your famous Turing Test has been a benchmark for AI. How do you think it aligns with today's AI capabilities?
GPT Turing: The Turing Test was less about passing a definitive threshold and more about exploring the possibilities of machine intelligence. Today's AI can simulate convincing conversations, but the test's deeper question remains: Can machines exhibit intelligent behavior indistinguishable from humans, not just in form but in understanding and consciousness? We're seeing remarkable strides in form, but the latter remains a frontier.
Reading the simulated interview between Steve and ‘GPT Turing’ broadened my perspective on how we can use artificial intelligence tools to understand technological advancements and innovations from different lenses or perspectives. We as humans have a tendency to gravitate towards emerging technologies and ‘innovative’ ideas. But when we dig deeper into our history, we see how today’s cutting-edge technologies are often grounded in foundations in the past.
Talentless CEO Steve Mudd takes this conversation ever farther in the latest episode of the AIAF podcast: “The 2023 AI year in review feat. Alan Turing and a big AI whistleblower’
🔗 https://open.spotify.com/episode/676SJW2al90L8FizlcJzHL?si=EttY-WISc-7XJ6xe5k1MQ