Today’s internet is stuck in a massive balancing act: how do we protect free speech while keeping fake news from further polarizing society? As more young people ditch traditional TV news and turn to their phones for daily updates, social media apps have taken on a huge responsibility for deciding what information spreads. Instead of deleting every post that seems fake, major apps are trying smarter, more subtle ways to handle bad info. To see how content moderation is changing, we can look at two of the biggest platforms on our phones: Meta (which owns Facebook, Instagram, and Threads) and TikTok. Both platforms have built very different strategies to fight digital harm while keeping online conversations open.
Meta has completely changed how it polices posts. For a long time, the company hired professional journalists and independent groups connected to the International Fact-Checking Network (IFCN) to review viral posts and mark them as false. Recently, Meta moved toward a crowdsourced system similar to X’s Community Notes, putting regular users in charge of giving context. Under this new system, instead of relying only on professional fact-checkers, users write helpful notes to add missing facts to posts. Another algorithm checks if people from different political backgrounds agree that a note is helpful before showing it to the public. While giant pop-up warnings have been scaled back in favor of smaller context boxes, advanced AI tools still automatically take down extreme dangers, like child harm, cyberbullying, and fake voting dates, a process heavily analyzed by digital research groups like the Stanford Internet Observatory.
Imagine someone posts a fake rumor about local voting rules or a fake medical cure. Under the old rules, a certified fact-checker would write an entire article explaining the lie, and Meta would hide the first image behind a dark warning screen that forced you to click through to read it, while silently hiding the post from search results. Under the new rules, the post stays up without a pop-up or block. Users who spot the rumor can add useful context notes linking to official government sites or sources like the World Health Organization (WHO). If Meta’s system sees that users with different viewpoints agree the note is fair, a small box appears directly under the post to give readers the full story without deleting the original post.
TikTok faces a different challenge because its app is made of short videos on the “For You” Page (FYP), where fake news spreads through edited audio clips and quick video edits rather than long text posts. TikTok relies on cutting off viral distribution, clearly labeling AI videos, and using educational strike systems for creators. When a video has unproven claims during a fast-moving disaster, TikTok rarely waits for users to write notes. Instead, it quietly removes the video from the FYP algorithm, preventing millions of people from seeing it. TikTok also upholds strict rules backed by guidelines from the Coalition for Content Provenance and Authenticity (C2PA), requiring creators to label realistic AI-generated media and banning fake videos that trick people about public safety or elections. To keep creators informed, TikTok uses a warning-strike system paired with an “Account Check” tool that lets users see whether their past videos broke any rules.
For instance, if a creator posts a fake video using an AI voice-clone of a government official declaring a fake national emergency, TikTok’s team acts immediately. Because the video uses sneaky AI edits to pose a public safety risk, it gets wiped from the app entirely. On the other hand, if a creator posts an unverified rumor about a sudden weather crisis, TikTok’s system keeps that video off the FYP, attaches an informational banner leading to official updates from safety groups like the Federal Emergency Management Agency (FEMA), and sends a warning notification to the creator’s dashboard so they learn the rules.
Deciding which platform has the better strategy means combining open debate with public safety. Meta’s move to crowdsourced notes addresses a common complaint highlighted by tech policy researchers at the Pew Research Center: that social media companies act like biased gatekeepers when deciding what is true. However, letting regular users write notes creates a huge speed problem. Misinformation spreads like wildfire in its first few hours. Waiting for users with opposing political views to agree on a note means millions of people might see a lie before a correction appears. Plus, on super controversial topics where people rarely agree, a note might never get approved at all, leaving massive lies totally uncorrected.
TikTok’s strategy of removing videos from recommendations addresses how fast short-form videos go viral: taking a video off the FYP instantly slows its reach. Its proactive stance on AI-generated audio is also crucial as deepfakes become easier to make. But the main downside of TikTok’s approach is its lack of transparency, a problem digital rights advocates at the Electronic Frontier Foundation (EFF) frequently raise. Because TikTok quietly restricts videos without deleting them, regular users often wonder why certain topics or political opinions suddenly vanish from their feeds. Even though creators get dashboard alerts, average viewers rarely know why certain trends disappear while others take over their FYP.
To actually slow down fake news without resorting to censorship, apps need to combine smart friction, digital literacy, and clear rules. First, apps should introduce “speed bumps” whenever a user tries to share a fast-trending claim that hasn’t been verified. A small pop-up asking, “Have you checked this with a second source?” encourages lateral reading, a fact-checking habit highlighted by the Stanford History Education Group (SHEG), by prompting users with a warning or advertising that the information could not be truthful. Apps using community notes like Meta could build a system in which universities or researchers review information and posts during major emergencies, helping publish accurate corrections when time matters for public safety. Platforms could use pop-ups to automatically mark any video without a verified tag as “Unverified Information,” so creators must prove their content is real.
Stopping online misinformation isn’t about picking a single company to decide what is true. Moderators raise concerns about unfair censorship, while secretly hiding videos can erode trust in what users see. Meta’s crowdsourced notes and TikTok’s algorithmic are two different paths for managing misinformation and how it can be censored online. By using transparent context tools with smart speed bumps that encourage critical thinking, social media platforms can help everyday users spot fake news for themselves and build a much healthier digital world.