AICE General Paper · Concrete Detail

The modern platform is not a library. A library chooses its books; a platform merely hosts what its users post, and the law has long treated that distinction as a shield. In the United States, Section 230 of the Communications Decency Act, passed in 1996, states that platforms are not “the publisher or speaker” of user content. That single sentence, written when Mark Zuckerberg was eleven years old, is the reason Facebook could grow from a Harvard dormitory project in 2004 into a company worth hundreds of billions. But the world the statute assumed no longer exists, and the damage it permits is now documented in courtrooms and in graveyards. Social media companies should be held responsible for the content they amplify, because their algorithms choose what to spread, their scale makes the harm global, and their failure to act has already cost lives. First, the distinction between hosting and amplifying is a legal fiction. When Facebook’s recommendation system promotes a post to a hundred million people, it is not passively storing it; it is choosing it. The clearest evidence came in 2018, when the Cambridge Analytica scandal revealed that the personal data of 87 million users had been harvested and used to target political advertising. The platform did not merely store that data; it enabled its abuse. In 2021, the whistleblower Frances Haugen testified before the United States Senate that Facebook’s own internal research showed its algorithms pushed users toward increasingly extreme content. If the company knew, and its internal documents say it did, then liability is not punishment; it is the ordinary cost of knowing. Second, the harm is not theoretical. In Myanmar, United Nations investigators concluded in 2018 that Facebook had played a “contributing role” in the violence against the Rohingya, because the platform let hate speech spread unchecked while the company employed almost no moderators who spoke Burmese. In March 2019, a gunman livestreamed the Christchurch mosque shootings in New Zealand, killing 51 people, and the video circulated for hours before it was removed. Each of these cases followed the same pattern: the platform learned of the danger too late, defended itself with the language of free expression, and promised afterward to do better. Admittedly, holding companies responsible raises real risks. Over-removal can silence legitimate speech, and a small startup cannot afford the moderation staff that a giant like Meta employs. Germany’s NetzDG, passed in 2017, fines platforms up to fifty million euros for failing to remove illegal hate speech, and critics warned that it would produce over-blocking. Yet the European Union’s Digital Services Act, agreed in 2022, sketches a more careful path: it requires the largest platforms to run independent audits and to open their data to researchers, without making a single company the judge of truth. Regulation can be precise; it simply has to be. The counterargument also fails on its own terms. Australia’s Online Safety Act of 2021 gave a regulator the power to order the takedown of abhorrent violent material, and the platforms complied. Ireland’s Data Protection Commission fined Meta 1.2 billion euros in May 2023 for mishandling user data, and Meta did not leave Europe. Companies adapt when the price of inaction is real. What they ignore is a fine they can absorb. The question, then, is not whether platforms should be responsible, but whether we will keep letting them be the only industry in history to profit from a product they refuse to stand behind. Section 230 was written to let a young internet grow. That internet has now grown up, and the shield has become the weapon. Responsibility is not censorship; it is the price of scale.

What each highlighted detail is doing