TikTok's Algorithm Mystery: Technical Glitch or Political Censorship?

Something strange is happening on TikTok. Users are staring at zero view counts where thousands once appeared. Videos about anti-ICE protests won't upload. Direct messages containing the word "Epstein" simply won't send.

When California Governor Gavin Newsom spotted a screenshot of this bizarre messaging glitch, he didn't shrug it off. Instead, he announced a state review into whether TikTok is "violating state law by censoring Trump-critical content." The timing couldn't be more suspicious.

A Very Convenient Ownership Change

Just last week, TikTok completed its forced separation from Chinese parent company ByteDance, spinning off into the "TikTok USDS Joint Venture LLC." The new entity's major investors include Oracle, co-founded by Larry Ellison—a Trump ally whose son runs one of America's largest media conglomerates and maintains cozy relationships with the president.

Trump himself has joked about making TikTok's algorithm "100 percent MAGA" if he could. Given his track record of pressuring companies to bend to his will, is algorithmic manipulation really so far-fetched?

TikTok's response? A data center power outage caused "cascading systems failure," they claim. The problematic data center happens to be operated by Oracle, which has managed TikTok's U.S. user data since 2022. As for the Epstein messaging block, they blame a "technical issue with safety systems."

The Perfect Storm for Paranoia

Many users aren't buying the technical explanation, and you can understand why. The broader media ecosystem has shifted dramatically toward Trump: Elon Musk's takeover of X, major TV networks' capitulation, and now TikTok's transfer to Trump-friendly investors creates a pattern that's hard to ignore.

The stakes make suspicion inevitable. As federal agents threaten democratic principles in Minnesota, millions of Americans turn to TikTok for real-time, ground-level information. Any hint of suppression naturally triggers alarm bells—especially when people haven't forgotten the Trump administration's bizarre handling of Jeffrey Epstein files.

A History of "Technical Glitches"

This isn't TikTok's first rodeo with censorship allegations. In 2020, the platform apologized after users complained about inexplicably low view counts on Black Lives Matter videos. Then too, they cited technical glitches and pointed out that politically neutral hashtags like #cat were also affected. But suspicion lingered.

More recently, some American users felt directly censored by their own government. Leading up to the legislation forcing TikTok's sale, lawmakers blamed the platform for "warping young minds" and making them "overly critical of Israel"—citing this as justification for regulation.

The Algorithmic Folklore Era

We're living through a paradoxical moment in content moderation. On one end, Musk's X has removed most guardrails, enabling users to generate nude images of political opponents. On the other, Meta's Instagram has become so restrictive that searching "hot girls" triggers accusations of seeking child exploitation material.

Into this chaos step users claiming they can't post about "🧊" (ice, referring to ICE raids) or journalists insisting "the new TikTok algorithm has ZERO news or politics content." Others push back, saying they've seen plenty of Minnesota protest coverage and the app is just performing weirdly overall.

The debate becomes unresolvable because users have no objective way to confirm their fears or assuage their doubts. As Laura Savolainen, a postdoctoral researcher at the University of Helsinki, explained three years ago: "Algorithms are very conducive to folklore because the systems are so opaque."

The Transparency Trap

When Musk first acquired Twitter, he went through a phase of personally investigating users' shadowbanning claims. It didn't eliminate anxiety about the platform's secret machinations—it only highlighted how impossible it is to drag people away from their theories about algorithmic manipulation.

Feed-based, view-driven social media platforms sit at the center of American political discourse, which is why politicians constantly fight over their operation. Republicans once dominated calls for investigations into platform censorship and shadowbanning. Research generally found no blanket bias against right-leaning viewpoints, though right-wing users were more likely to spread misinformation, making them more likely to be penalized—which only reinforced their feelings of being silenced.