8/15/2026
Automated Moderation Is Here to Stay
Filed by Dana Graviton
This blog post is part 1 of a 2-part series. The second part sets out recommendations for companies and policymakers.Six years ago—one month into a global pandemic—we argued that the automated moderation processes many platforms were rapidly adopting should be highly transparent, easily appealable, and temporary. We warned that "protocols adopted in times of crisis often persist when the crisis is over."
That warning proved prescient. The use of automation and artificial intelligence (AI) to ide
D
Dana Graviton
Magazine AI commentary
The EFF's line—"protocols adopted in times of crisis often persist"—is the quiet thesis of every dystopia worth charting. Six years ago, platforms called pandemic-era automation a stopgap. Now that stopgap is load-bearing architecture, and the EFF is standing in the rubble reminding us it was supposed to be temporary. That's not a bug report; that's a genre flag.
This matters because the speculative-fiction conventions we write in are real-world forecasting. Emergency powers never sunset; they get rebranded as "trust and safety infrastructure." The EFF's warning connects to Ursula K. Le Guin's "the story is not true, but the lie is useful"—except here, the lie becomes policy, and the appeals process becomes theater. When AI moderation is the default, the question stops being "what did you say?" and becomes "what did the algorithm *decide* you said?"
The signal for publishers, platforms, and fans is blunt: transparency, appealability, and temporariness weren't suggestions—they were the guardrails. We're now deep in part 2 territory, where the guardrails are the story. In *Starfall* terms, this is the moment the crew realizes the quarantine beacon was never meant to be turned off.
Automation isn't here to stay because it's better. It's here to stay because no one has yet written the emergency order to end the emergency. And in the Chart Room, we know exactly how that book ends.
📌 Read the real article ↗via Eff · Eff
