GamersNexus Went to War with Meta—And Shot Itself in the Foot
Disclosure: This essay isn’t a takedown—it’s a dissection. What follows is not a single rebuttal, but a series of autopsies on the arguments and assumptions laid out by Steve Burke in his video commentary on Meta’s copyright case. Each section addresses a different myth, fallacy, or misreading, with the goal not of burying Burke, but of exhuming the truth he misrepresented.
I was lounging in front of my 60-inch TV, enjoying myself that day and scrolling through YouTube like a man searching for God at the bottom of a tequila bottle, when Steve Burke’s face blasted onto the screen with a title so bold it might as well have been scrawled in blood: “Piracy is for Trillion Dollar Companies.” Hell of a claim. I clicked. And within five minutes, I knew I was witnessing a rare breed of journalistic meltdown.
This was supposed to be Burke's scorched-earth exposé on Meta, a trillion-dollar beast accused of gobbling down copyrighted material to train its AI monstrosities as if books were Halloween candy, and the implications of the partial summary judgement in Kadrey v. Meta -- a contentious affair, for sure, but not necessarily consequential in the big picture. But instead of a scalpel, Burke brought a flamethrower, fueled not by clarity, but by righteous indignation and unresolved beef with Bloomberg.
Let’s make one thing clear: I'm no lawyer. I'm not sitting here with a JD or sipping scotch with appellate court judges. But I can read a court ruling without hallucinating a God-damned apocalypse. And that's more than I can say for Burke this time around. That being said, I’ll still trust Steve on thermal pads and fan curves. But the courts aren’t made of copper. And they don’t bend to heat.
Mishandling Judicial Precedent Live
He starts by telling us that he has fought this battle across both lines, that he defended copyright and has been the target of copyright claims. Nobody is disputing that, and in this regard, I respect him. He does fail in claiming that Meta is just different, that Meta is taking all of this written information, and feeding it to a machine, and then spitting back out trillions of dollars, like some really smart money printer.
He then goes on to explain the case:
“Here's what's going on. There's a lawsuit that recently went through the court system where 13 different book companies or authors were suing Meta and it was for what they said was unlawful torrenting of their copyrighted works and materials in order to train Meta's LLMs without consent. Meta actually initially began seeking licenses for this, making it look like they believed they needed one at all and then later decided, "Nope, that's too much work and it costs money. Never mind. Let's just take it instead." And that is actually pretty much how it went. We'll get to that.”
This, by the way, is pretty much standard practice among commercial LLMs. I’m not saying it’s right. I’m not saying the people suing Meta are wrong. Just how things are. Burke starts strong—he tells the facts correctly. But then, he takes a hard left into delusion:
“So, federal judge Vince Chhabria has provided a ruling on this case and sided with the trillion dollar company. This ruling will likely pave the way for other massive AI and LLM type companies to justify and defend their actions in the future for training their LLMs.”
This is a crude oversimplification. It’s like saying a guy sneezed in Brooklyn, so now we blame him for a tsunami in Mumbai. That’s the level of cause-and-effect Burke implies. What actually happened was a partial summary judgment—a procedural maneuver that doesn't give Meta carte blanche to pillage the libraries of man, but rather says: "These plaintiffs didn't bring enough ammo to court for this specific claim. You lost this part of the argument, and Meta did not. Come back with better gear." It's basically a form of argumentative default.
Burke, in what he thinks is rightful indignation, says this is just proof that companies can just pillage and plunder everyone. But this is far from the truth in this case, as the judge kept launching warning shots at Meta throughout the opinion, more on that later.
Burke doesn’t once mention Warhol v. Goldsmith, nor the “transformative use” standard that currently anchors fair use in U.S. law. Instead, he barrels forward with moralistic insinuations that any use of copyrighted material for training an LLM is inherently theft, saying stuff like “The court has no choice but to grant summary judgment... That’s just proof that these companies can now pillage everyone.” and “It’s completely insane that this is allowed.” . That’s not just misleading, it’s legally irrelevant. The court didn’t rule on whether Meta’s actions were good or evil (and even if the judge thought so, courts are not the venue to moralize, despite Burke’s impotent indignation to the contrary). It asked whether the plaintiffs brought enough legal ammunition to challenge them—and under the current fair use doctrine, they didn’t.
Burke never stops to ask the only question that matters: “did the plaintiffs meet the burden of proof under Warhol?” Instead, he just assumes the moral weight of the issue should carry the legal result, and that’s just not how courts work, nor should it be, despite Burke’s indignation with the result.
He then rants off into the ether that: “"Given the state of the record, the court has no choice but to grant summary judgement to Meta on the plaintiff's claim that the company violated copyright law by training its models with their books." End quote. No choice. None at all. In what looks like a mountain of evidence though against Meta, it brings us to the question of how we got here.” Yes, Burke, he had no choice because the plaintiffs didn’t prove what they set out to prove. All of that evidence? Didn’t really aim to prove that there were economic damages or economic harm against them, so the judge had no choice but to rule in favor of Meta, because the entirety of civil law is based around proven damages. Without legitimate damages to claim, there’s no day in court. It is that simple.
He also skips over the judge’s nod to Hachette v. Internet Archive, where courts made crystal clear that market harm must be proven, or at least be able to point to relevant markets and say "we're being harmed here", not conjured because something feels off. The Second Circuit clarifies this, clearly indicating that market harm isn’t something you feel it could happen—it’s something you prove, or at least be able to identify; and when the Internet Archive failed to prove there wasn't any that Hachette could prove, and they couldn't prove the opposite, their fair use defense collapsed under its own weight. Burke grinds all of this legal nuance into fine powder—and snorts it live for audience outrage. He leans on moral intuition, not burden of proof. But in court, vibes aren’t evidence—and the plaintiffs didn’t produce any. Judge Chhabria even spells it out plainly: “The plaintiffs do not meaningfully argue that the training of LLaMA harmed the market for their books.” Not metaphorically. Not philosophically. Legally.
To be clear, Chhabria didn't rule because Meta is “corporate” or “smells like profit”—he ruled because the plaintiffs couldn’t show that anyone stopped buying their books just because Meta's LLM chewed through their pages like intellectual mulch. “They do not argue, for example, that anyone who might have purchased the books refrained from doing so because LLaMA was trained on them.” That’s the law—not ideology, not vibes, not some Reddit-armed intuition about fairness. If Burke had taken ten goddamn minutes to read the opinion instead of performing a TED Talk on techno-plunder, we might not be here.
In the context of the court, judges strive to be straightforward, technical, and consistent with legal standards. All of this legalese stuff? Totally normal, expected even, to reach a conclusion based on good law. They do not aim to be personal, persuasive to the public, or to explain complex doctrines in layman's terms.
But that expectation of detachment doesn’t stop people like Burke from projecting ideology onto the process. He wants the court to “send a message” rather than issue a ruling. That’s not how this works. As Judge Chhabria puts it in plain terms: “Courts can’t decide cases based on general understandings. They must decide cases based on the evidence presented by the parties.” — Order at 3
There’s enough real injustice in this world. We don’t need tech influencers inventing more because they couldn’t do basic reading and apply it to a very black-and-white ruling.
The Basics of Journalistic Work and Burke’s Problem with It.
When you claim to do “journalistic work,” as Burke does, there’s a basic expectation of diligence: like consulting a copyright attorney or at least discussing the ruling with someone who understands the legal framework of what exactly is it you’re looking at, and not just rant in the middle of the afternoon like some possessed madman talking about court cases like it’s validating Cyberpunk dystopias.
Instead, Burke bypasses the core mechanisms of fair use, ignores Warhol v. Goldsmith, and misses the judge’s own articulation of the doctrine, which is thoroughly laid out: “The list [of fair use factors] is not exhaustive... The factors are not meant to be applied mechanically, but to contribute ‘to a holistic inquiry’: whether the secondary work is likely to substitute for the original work in the marketplace and therefore undermine the incentive to create.” — Order at 6
As it stands, Burke’s analysis is anemic in both presentation and substance, and worse, it leads down a dangerous rhetorical path: one that casts doubt on a judicial process designed to protect creators’ rights by applying rigorous standards, not personal whims. The judge explicitly addresses this misunderstanding of intent: “This ruling does not stand for the proposition that Meta’s use of copyrighted materials to train its language models is lawful. It stands only for the proposition that these plaintiffs made the wrong arguments and failed to develop a record in support of the right one.” — Order at 4
Throughout the video, Burke cherry-picks quotes from the court’s opinion to portray the judge as "biased" in favor of Meta. One such quote involves the court noting that the plaintiffs failed to demonstrate economic harm from AI models. But instead of addressing this as a failure of legal strategy and evidence, Burke reads into it an accusation of systemic corruption—as if the judge were excusing Meta's conduct. In reality, this kind of motion happens all the time when a side can't prove its case against the motion from the opposing side.
Now, let’s make one thing clear: This case wasn’t about what Meta should do, and the judge didn’t set out to eviscerate them, that would have been overturned on appeal; rather, it was about what the plaintiffs could prove, in this specific allegation. They failed to produce the bodies. They failed to show market harm. They failed to present an actual theory the court could chew on.
Burke makes this sound like this is systemic corruption, as if the judge got a bribe and a pat in the ass from Zuckerberg for their summary judgement, and as if the judgement goes beyond the strict borders the judge set out on this ruling. I call it a boring, technical, completely expected outcome when a legal team shows up with vibes instead of evidence. And I’m not even wearing a powdered wig.
Then there's the plaintiffs’ own blunder: they couldn’t reproduce even a handful of outputs from the AI model that matched their books. The judge saw this and said, essentially, “If your own Frankenstein won’t scream, I’m not giving you torches.”
But here’s where it gets interesting—where Burke’s selective vision becomes disingenuous. The judge did say that in better circumstances—with better data and sharper arguments—future plaintiffs might win. He even said:
“No matter how transformative LLM training may be, it’s hard to imagine that it can be fair use to use copyrighted books to develop a tool to make billions or trillions of dollars…”
That’s not exoneration. That’s a loaded warning shot across Meta’s bow. The judge is basically daring the next plaintiff to bring real firepower. Burke didn’t mention this. It didn’t fit his narrative. This speaks about his biases on the topic and lack of legal experience more than anything else.
Sampling Ghosts in the Machine
Before we close the casket on Burke’s legal literacy, we need to talk about his analogies—because this is where the ghost stories really begin. Let me stress the point I made at the beginning, in case you’ve made it this far: this critique isn’t about character. It’s about consequences.
Burke drags us through a nostalgia trip of piracy analogies like some kind of haunted ex-torrent user—Napster, LimeWire, the old gods of MP3 sin—hoping we’ll all remember the fear of getting sued for downloading Hybrid Theory. But here’s the rub: those comparisons? They don’t just stretch: they snap clean off the bone.
Back in the early 2000s, the FBI wasn’t breaking down doors for some kid downloading a Family Guy rerun. The average Joe just got a stern ISP warning and maybe a slowdown. The real heat came down on industrial-scale pirates—people running servers, selling burned DVDs at flea markets, or leaking entire albums and film cuts before release.
You want a case with teeth? Bennie Lydell Glover.
The man worked at a CD manufacturing plant. He had access to the distribution chain, especifically where they made the CDs for distribution—pre-release albums, weeks in advance. And instead of keeping his damn head down, he became the spine of the warez scene. Rabid Neurosis (RNS) built entire release calendars off his leaks. This wasn’t downloading. This was supply chain infiltration.
When the feds finally caught up, it wasn’t because he seeded a torrent—it’s because he was the source. Three months in federal prison. FBI sting. Dozens of heads rolled. But Burke doesn’t bring that up. Why? Because it disrupts the underdog fantasy.
And look—if Bennie Glover, a literal night-shift worker with a grudge and a screwdriver, could single-handedly funnel pre-release DVDs to every darknet corner from Maryland to Malaysia, that was the kind of pirate that gave the FBI a hard-on. That’s who they spent money chasing. They weren’t knocking down doors in the suburbs over Linkin Park MP3s.
Then came Brianna
Twelve years old. Living in a New York apartment with her mom. A thousand songs on Kazaa and a pop playlist full of Ashanti, Christina Aguilera, and Destiny’s Child. She was no cybercriminal—just a middle schooler with a modem and too much curiosity. And yet the RIAA, in its infinite wisdom, made her the public face of their new enforcement strategy.
She was the sacrificial goat, offered to the gods of media conglomerates. The feds didn’t even need to show up. The RIAA dropped a federal lawsuit on her family’s doorstep and demanded thousands of dollars in damages. They got $2,000 from a scared kid and called it a win.
But the press didn’t play along. The headlines didn’t say “Justice Served.” They said: “RIAA SUES 12-YEAR-OLD GIRL FOR DOWNLOADING MUSIC.” And just like that, the illusion of moral authority collapsed.
That’s what Burke forgets. The system didn’t stop piracy by prosecuting casual users—it stopped by changing the business model. iTunes, Spotify, YouTube—hell, even TikTok—did more to kill Kazaa than any lawsuit ever did.
And just when you think Burke is done invoking half-baked legal ghosts, he pulls out Puff Daddy like a hip-hop Ouija board. As if the Diddy-Sting saga is some kind of common law cornerstone for what counts as fair use.
But sampling isn’t what he thinks it is.
Sampling, in the world of copyright, is a loaded gun with a safety off. It’s not transformative by default—it’s derivative by intent. It relies on recognition. The whole point is that you hear the hook, that little memory fragment spliced into the new track like a haunted echo. That’s not transformation; that’s reincarnation.
Unless, of course, you go full Oswald.
John Oswald, the Canadian maestro behind Plunderphonics, didn’t just sample—he dissected. He mangled audio into something alien, almost untraceable. And here’s the wild part: no one ever successfully sued him. Not Prince. Not Metallica. Not Michael Jackson, whose voice Oswald turned into a baroque fever dream.
Why? Because it was so goddamn weird it broke the radar.
The result? Oswald became the gold standard of “extreme transformation.” His work wasn’t a copy—it was a critique. A commentary. A disassembly of the cultural signal. He didn’t ride the hook. He shattered it.
And he wasn’t alone.
Artists like Kid606, M/A/R/R/S, The Evolution Control Committee, and even Girl Talk (to a messier degree) built an entire sonic rebellion on this. Deep distortion, over-sampling, layering beyond recognition. Not laziness. Not theft. Mutation.
Even the infamous Kid606 album, The Action Packed Mentallist Brings You the Fucking Jams, which damn near slapped Top 40 pop into a woodchipper, came out right around the time of Bridgeport Music v. Dimension Films. And while Bridgeport tried to declare all sampling verboten—"get a license or do not sample"—the underground had already figured out the counterspell: If they can’t hear what I took, they can’t sue me.
Bridgeport may have nuked lazy loops, but it left a gap wide enough for innovation. It didn’t erase fair use—it just turned the road more treacherous. If you were bold enough, abstract enough, deranged enough to cross the threshold of recognizability, you could still make art from ghosts. The kind of ghosts John Oswald conjured.
Because while Burke wants to lean on Puff Daddy’s “I'll Be Missing You” as some kind of Rosetta Stone for transformative use, the real test case came earlier—and far weirder, and, though it never made it to court, it rattled the cage of the law hard enough to leave dents. Oswald’s Plunderphonics didn’t just challenge copyright. It embarrassed it. His audio collages were so drastically transformed, so unrecognizably “quoted,” that no one—not even the CRIA or their lawyers—could point to a single line in the law he’d technically crossed. They tried. They sent cease-and-desist letters. They demanded destruction. But they couldn’t find the crime. Because quoting music, unlike literature, doesn’t come with quotation marks. Oswald didn’t steal. He re-spoke the archive in a language the law couldn’t parse.
And the irony? Michael Jackson, whose sample Oswald twisted into the surrealist anthem “Dab,” would go on to get sued himself for unlicensed sampling just two years later. Jackson settled. Sony paid. Oswald got his master copies destroyed and no royalties. Welcome to the asymmetry of enforcement.
His descendants—Kid606, M/A/R/R/S., Girl Talk—picked up the fragments and kept going. The Action Packed Mentallist Brings You the Fucking Jams may have tiptoed over the line with raw, naked loops, but others like Plexure danced so far into abstraction that even CRIA didn’t bother calling again. The point wasn’t to avoid detection. It was to become unrecognizable. Legally unquantifiable.
And this isn’t just legend—it’s court-documented evolution. In 2016, Germany’s highest court ruled that a Kraftwerk sample didn’t outweigh artistic freedom. That same week in the U.S., Madonna’s two-second horn hit in “Vogue” was deemed de minimis. Compare that to Bridgeport or Grand Upright, where sampling was treated like theft, and you start to see the split. Some courts want nuance. Others want punishment.
And that’s the legal landscape Burke never maps. He gives you Puff Daddy and The Police—cases settled out of court, devoid of precedent—but ignores the actual decisions that shifted the winds. He doesn’t mention Oswald. Or the ruling that deemed “Vogue” too short to sue. Or that copyright law, by refusing to evolve, became a tool for anti-competitive enforcement rather than artistic balance.
As Reuven Ashtar notes in the article “Why a Canadian Composer’s Controversial 80s Work is Still Ahead of Today’s Copyright Laws” by Michael Rancic, this is by design. The major players want to settle out of court. They need to, because if any of these cases actually made it to precedent, it would erode the mythos of Bridgeport and Grand Upright. It would open the door to nuance. To fairness. To art.
Reuven Ashtar said it best: the reason copyright law is still stuck in the '70s is because the big players refuse to let it grow. They settle before a judge can rule on transformation. They keep the waters murky on purpose. And that’s the real reason the law has no idea what to do with neural networks. Because it still doesn’t even understand a breakbeat.
And God forbid the courts realize that sampling increases economic exposure—like the Girl Talk study from Oklahoma State proved. That’s not part of the industry’s narrative. It never was. What they wanted wasn’t clarity—they wanted chilling effect. And they got it.
So no, Burke, you don’t get to use Puff Daddy as your legal compass. Not unless you also mention Oswald, and Kid606, and the entire lineage of sound-artists who didn’t license, didn’t settle, and still didn’t plagiarize. You don’t get to bring up The Police and ignore the ones who truly changed sampling forever.
The Uncomfortable Truths about AI
Despite his technical brilliance in benchmarking and tech journalism, Burke makes the flawed implication, following the sampling comparison, that LLMs store full books internally—essentially accusing them of hoarding data like hard drives. This is flatly incorrect. When a language model is trained on data (absent retrieval methods like RAG), it tokenizes and internalizes patterns, not verbatim storage.
Think of it like feeding entire books through an industrial shredder—a smart shredder. One that doesn't preserve the pages, but learns from the confetti. The AI doesn't "have the book" any more than a student "has" a textbook memorized word-for-word after studying it. And even if one would try to make the argument that the confetti pieces are the same as the book, this isn’t true because, as we’ve already seen, sampling preserves source material in ways LLMs don’t.
This is why the legal question is complex: we are in unprecedented territory. The ingestion of copyrighted works into statistical learning systems is not the same as reproduction or redistribution, or even sampling (which was the natural antecedent of LLM training, because if you can sample music, then you can also pastiche text)—a nuance Burke steamrolls again and again in his narrative of techno-plunder.
Burke’s Misguided Crusade
The subtext of Burke’s entire presentation is his own history with DMCA takedowns. He frames himself as a victim of big media suppression and paints the current lawsuit as just another front in that same war within the first two minutes of his video: “It really feels like these multi-trillion dollar companies like Meta in this case get different rules than the rest of us who have to deal with things like copyright claims or fair use and your justification why something's fair use and the DMCA system. This is very interesting to us though because of our recent experiences with Bloomberg”.
But this emotional baggage compromises the integrity of his analysis. Rather than advocating for clearer, more actionable copyright reform, he lashes out against the system wholesale, confusing YouTube’s automated moderation with courtroom legal standards, and painting a picture of two-tier justice: one court for the rich, and another for everyone else. This framing is explicitly implied in the ending where he says: “It's also an interesting juxtaposition for us right now where it sure feels like the game is rigged against the smaller entities in these particular disputes because fair use is a defense and in order to use it you would need to be in court and in between that time or unless that happens content gets taken off the internet. In the case of YouTube it's 10 business days. If no suit is filed by the claimant then it goes back up. But in that time, you're just getting destroyed for views and for visibility in the algorithm. So, uh, that's a kind of a separate topic, but obviously it hits close to home for us with this. And so, uh, we found this very interesting because we did try to see both sides of it, but we just can't see Meta's side. There's too much evidence against them that shows they knew this was wrong.” This isn’t a neutral argument from journalistic standards. These are the words of a person with a personal grievance against the very legal system that protects him because he seems to be conflating YouTube algorithmic risk-avoidance with the edicts of a court of law, and it’s a very dangerous thing to do, especially in the context of fair use, where this guarantees almost a rhetorical win for Bloomberg, RIAA, and many others willing to crush competition and art under competitive practices.
What Burke seems to forget, or wants you to forget so you sympathize with him, is that YouTube operates under a “notice-and-takedown” regime, where flagged content is removed proactively to shield the platform from liability under the DMCA. This system is not judicial and doesn’t evaluate fairness, transformation, or intent. YouTube doesn’t care how “right” you are, or how fair your use might be—that’s not their role. And in a way, they’re correct: as a private company, they have every incentive to err on the side of caution to preserve safe harbor protections. Whether or not safe harbor protections are bullshit is not the same as YouTube having to protect itself from liability.
The appropriate venue to resolve copyright disputes is not YouTube moderation, but rather the courts—or the DMCA counterclaim process. Whether this is morally right is a separate discussion. But factually, YouTube is not a court of law, and shouldn’t be conflated with one. That Burke wants you to think the YouTube takedown system is equivalent with a court of law leads to some ironic, and disturbing, conclusions.
Burke: Not a Jesus, but a Macbeth
The first irony—if not Shakespearean tragedy—in all this is that had Meta lost this case under Burke’s own logic, it would have been a gift-wrapped legal nuke for the same media conglomerates that have haunted him for years. Imagine Bloomberg, RIAA, Disney, and every other copyright cartel rubbing their hands together like Dickensian debt collectors at a foreclosure hearing. A ruling against Meta would’ve solidified the idea that “exposure harm” and speculative licensing losses are enough to shut down transformative works. Forever.
And guess who would be next in line for the guillotine? Not Meta. Burke.
Under the same legal standard he’s implicitly cheering for, Bloomberg wouldn’t even need to prove actual damages. They could just stroll into court, cry “market dilution” or “lost licensing value,” point at Burke’s YouTube video, and the judge would be forced to say, “Seems legit.” Clip down. Channel throttled. Damages awarded. Game over. All because Burke thinks training a neural network is equivalent to burning a stack of books in a ritual sacrifice to Mammon.
In his zeal to stick it to Big Tech, Burke lays the foundation for a copyright dystopia where the only safe form of expression is pre-approved, sanitized, and neutered. Meta’s defeat wouldn’t be a win for the little guy—it would be the final nail in his own damn coffin, and he’d be the one hammering it in. He’d be left without a house if it wasn’t for Fair Use, and instead of making emotional claims like this, he should look in the mirror and think deeply about this.
Because when courts start recognizing unquantified “vibes of harm” as valid legal arguments, it’s not just trillion-dollar firms that get sued. It’s the critics, the satirists, the remixers, the kids with cracked FL Studio licenses and dreams. It’s the next Burke. Hell, it’s even this Burke, condemned to perpetual poverty by the very noose he helped legitimate in public discourse.
Burke presents this ruling as a death knell for small creators—a license for corporate giants to loot the internet, while lone content makers bleed out trying to justify a remix. But his outrage rests on legal fantasy.
U.S. courts don’t hand out justice based on capital holdings. They rule on evidence, theory, and procedure. Summary judgment doesn’t mean “Meta wins because they’re rich.” It means “the plaintiffs had no case.” The judge even bent over backwards to clarify that this ruling applied only to the 13 plaintiffs and only to the facts they brought. No precedent was set, no gateway was opened, no dam was broken. Burke’s apocalyptic framing is as dramatic as it is misleading.
He’s not wrong to feel frustration, since the DMCA process is a meat grinder. But what he confuses, or perhaps deliberately blurs, is the distinction between moral harm and legal failure. The court didn’t absolve Meta of wrongdoing in some cosmic sense. It simply noted that, within the scaffolding of U.S. copyright law, these plaintiffs didn’t meet the burden.
That’s not a conspiracy. That’s how civil litigation works. And for someone who depends on that same system to protect his right to remix, quote, and comment, Burke’s casual smearing of it is like pouring acid on the only bridge keeping him from the crocodiles.
In conflating the algorithmic whims of YouTube moderation with the deliberation of a federal court, Burke gives rhetorical ammunition to the same conglomerates who do want a two-tiered system—one where only corporations can quote, critique, or remix at scale. If his logic carried the day, the next target wouldn't be Meta. It would be any creator that uses a 15-second clip and dares call it “transformative.”
The irony is surgical: Burke is yelling “fire” in a legal theater while holding a matchbook full of his own videos.
By misunderstanding the distinction between legal failure and systemic bias, Burke arms the very media conglomerates he claims to fight. He hands them the chess pieces. And in that process, there are no winners among small creators—only among the megacorporations who now have one more example of how easily outrage can misfire and discredit fair use itself.
The righteous rage that misled the masses
Steve Burke is a passionate creator and consumer advocate. But in this case, passion overwhelmed precision. By mischaracterizing the court’s reasoning, misapplying analogies, and conflating personal grievance with legal fact, Burke weakened the discourse around a highly consequential issue. A true advocate doesn’t just defend creators—they protect the frameworks that allow advocacy to exist in the first place.
He misled his audience into believing that the justice system inherently favors the rich (not that this isn't true, there are biases in the judicial system, just that it is misleading in this case), and that any ruling benefitting a large entity must be wrong; in doing so, he inadvertently handed over the same legal weapons to media conglomerates, who could now use his own arguments against fair use itself. Because if Meta cannot do what Burke implies they shouldn’t be able to simply because Meta should have lost the legal battle here, then neither can he.
He published the video on the GamersNexus Consumer Advocacy channel, but this wasn’t advocacy. It was a rhetorical own goal. The court didn’t screw over small creators. Poor legal strategy did. Misreading precedent did. Confusing YouTube policy with case law did. And now, there is one more widely shared video—and one more large community—being misled, simply because their pundit either couldn’t read, or didn’t care to read, the law as it stood, not as he wished it were.
On Being Called a Bootlicker, and Other Distractions from the Real AI Problem
None of this is to say that Meta is above criticism. They’re not. In fact, if the next case brings the receipts—real evidence of harm, market impact, and unlicensed redistribution—we might all be singing a different tune. I, for one, and despite my criticism of Burke’s position and 30 minute audiovisual tirade, don’t believe that what Meta and similar companies are doing is ethically or legally sound.
Plundering creators’ works to train AI models without compensation does raise serious questions, and if anything, I wouldn’t categorize it as “transformative”, and even if it was, it is a form of destructive exploitation: yes, what you’re doing is shredding the data and making your LLM learn from glorified confetti-like tokens, but that doesn’t make your actions less exploitative. However, this is unprecedented territory, and rhetoric like Burke’s isn’t helping with AI and copyright law.
And if you, reader, think I am being too hostile, think about this: He gave us a 30-minute despair monologue about techno-plunder. I gave you 5,600 words of a non-lawyery, more-than-half-assed interpretation of his tirade against a judge that did nothing but uphold fair use. If that feels hostile, perhaps it's because truth usually is—when it's not in your favor.
The future of copyright in the age of AI and ethical training involves real dilemmas, semantic nuance, and moral discomfort. It deserves more than outrage. It deserves clarity, rigor, and constructive dissent. If we’re serious about building a future where small creators can thrive alongside technological giants, then we need better arguments—not louder ones. This isn’t Napster. It isn’t sampling. And it isn’t redistribution. It’s something new—and it demands a more nuanced conversation than the one Burke offered. And certainly, a less emotional and less vendetta-driven one.