What Is Section 230? The 26-Word Law That Built the Internet — and Why Everyone Wants to Change It
Section 230 of the Communications Decency Act is one of the most consequential laws in the history of the inar(--text-light);font-size:1rem;max-width:640px;margin:0;"> Section 230 of the Communications Decency Act is one of the most consequential laws in the history of the internet. Written in 1996, it shields online platforms from legal liability for content their users post and from lawsuits over their moderation decisions. Democrats want to strip protection for algorithmically amplified harms. Republicans want to end protection for politically biased censorship. Both want reform — for entirely opposite reasons.
- Section 230 (26 words) shields platforms from legal liability for user-posted content and for good-faith content moderation — it's why Facebook, YouTube, and X can exist without being sued for everything their users post
- Republicans want to reform it because they believe platforms use moderation immunity to censor conservative speech without accountability — they want to condition immunity on ideological neutrality
- Democrats want to reform it for the opposite reason — they argue platforms should face liability when their algorithms amplify harmful content (misinformation, extremism, child safety), not just passively host it
- Despite bipartisan reform pressure, Section 230 has survived intact because any majority coalition to change it fractures — Republicans and Democrats want incompatible changes
The 26 Words That Built the Internet
The core of Section 230 is this sentence: "No provider or user of an interactive computer service shall be treated as the publisher or speaker of any information provided by another information content provider."
This sentence — 26 words — means that websites, social media platforms, forums, and apps cannot be sued as if they were publishers of content that their users create. A newspaper that publishes a defamatory article can be sued for defamation. Facebook cannot be sued for defamation based on what its users post, because Section 230 says it is not legally the "publisher" of that content.
The law was a direct response to a 1995 court case, Stratton Oakmont v. Prodigy, in which a court held that because Prodigy moderated some content on its bulletin boards, it should be treated as a publisher and held liable for all content — including defamatory posts it had not reviewed. Congress concluded that this ruling would either force platforms to moderate nothing (to avoid publisher status) or moderate everything (an impossible task). Section 230 resolved the dilemma by decoupling moderation from liability.
Section 230(c)(2) adds a second protection: platforms cannot be held liable for "any action voluntarily taken in good faith to restrict access to or availability of material that the provider or user considers to be obscene, lewd, lascivious, filthy, excessively violent, harassing, or otherwise objectionable." This covers content moderation decisions — banning users, removing posts, labeling content — even when those decisions are controversial.
What Section 230 Does Not Cover
Section 230 is often mischaracterized as blanket immunity for any online content. It has significant limitations:
- Federal criminal law: Section 230 explicitly does not immunize platforms from federal criminal liability or prosecution.
- Intellectual property: Copyright and trademark claims are not covered — the DMCA (Digital Millennium Copyright Act) handles copyright separately with its own notice-and-takedown regime.
- Sex trafficking (FOSTA-SESTA, 2018): Congress created a carve-out allowing civil claims and state criminal prosecutions related to sex trafficking. Backpage.com was the direct target. Critics argue it drove sex workers underground and made them less safe.
- First-party content: If a platform creates content itself — rather than just hosting user content — Section 230 does not protect it. Platforms that pay for or actively solicit specific harmful content could lose protection.
- Antitrust: Section 230 provides no immunity for antitrust violations.
Reform Proposals: Two Parties, Opposite Goals
| Approach | Proponents | Core Argument | Key Proposals |
|---|---|---|---|
| Anti-censorship reform | Most Republicans | Platforms censor conservatives; immunity should require neutrality | TRUST Act, state anti-deplatforming laws (FL SB 7072, TX HB 20) |
| Algorithmic liability | Many Democrats | Recommendation algorithms cause unique harms; platforms should be liable for amplification | SAFE TECH Act, EARN IT Act, Algorithmic Accountability Act |
| Child safety carve-out | Bipartisan | Platforms should not be immune for child exploitation material | EARN IT Act (contentious), KOSA (Kids Online Safety Act) |
| Full repeal | Fringe voices both sides | Fundamental recalibration needed | No major legislation has proposed full repeal |
The Supreme Court considered the algorithmic liability question directly in Gonzalez v. Google (2023), which asked whether YouTube's algorithmic recommendations of ISIS content lost Section 230 protection. The Court declined to decide the merits, instead ruling narrowly on other grounds — leaving the algorithmic liability question open.
Frequently Asked Questions
Does Section 230 require platforms to be politically neutral?
No. Section 230 does not require any ideological neutrality as a condition of immunity. The law explicitly protects platforms' right to moderate content they find "objectionable" — without defining what counts as objectionable. Platforms can and do make moderation decisions based on their own policies, and those policies have been challenged as politically biased by conservatives. The Supreme Court held in Moody v. NetChoice (2024) that state laws requiring social media platforms to host all speech likely violate the First Amendment, because platforms have their own First Amendment rights to make editorial decisions.
How does Section 230 apply outside the United States?
Section 230 is a US law and applies in US courts. It does not protect platforms in foreign jurisdictions. The European Union's Digital Services Act (DSA), which took full effect in 2024, takes a different approach: large platforms (over 45 million EU users) must conduct risk assessments, provide algorithmic transparency, and remove illegal content, with significant fines for non-compliance. The EU model is closer to the "duty of care" approach that Democratic reformers in the US have proposed. Australia, UK, and Canada have all moved toward more regulated platform liability regimes than Section 230 permits.
What happened with the state anti-deplatforming laws?
Florida and Texas both passed laws in 2021 prohibiting large social media platforms from deplatforming political candidates or banning users based on political viewpoint. Both laws were immediately challenged in court. In Moody v. NetChoice (2024), the Supreme Court vacated lower court rulings and sent the cases back for reconsideration, instructing courts to conduct a more thorough First Amendment analysis. The Court indicated strong skepticism that states can compel platforms to carry content against their editorial judgment, without definitively striking down the laws. The cases were still working through lower courts in 2026.
