If you've clicked a Twitter link in the past couple of years and been greeted by a sign-in prompt instead of the tweet you wanted, you've met the login wall. It wasn't always like this. For most of Twitter's history, the platform was one of the most open large websites on the internet — anyone could read almost anything without an account. Understanding how that changed explains a lot about the modern web, not just about one platform.
The open era: Twitter as public infrastructure
From its founding in 2006 through the early 2020s, Twitter positioned itself as a public square, and its technical choices matched the rhetoric. Tweets were readable by anyone, embeddable anywhere, and accessible through a famously permissive API. An entire ecosystem grew on top of that openness: third-party clients, research datasets, embedded timelines on news sites, bots that posted earthquake alerts, and archives used by journalists and historians. Courts and governments treated tweets as public statements precisely because they were, in fact, public.
2023: the year the gates closed
The turning point came after the company's change of ownership in late 2022, driven by two connected pressures: the business needed revenue, and public tweet data had quietly become one of the most valuable AI training resources on the internet.
- February 2023 — the API price shock. Twitter announced the end of free API access. Academic researchers, hobby developers, and small tools that had relied on free or cheap tiers were priced out almost overnight, with enterprise access reportedly starting at tens of thousands of dollars per month.
- June–July 2023 — login required, then rate limits. The platform briefly required login to view any tweet at all, followed by the infamous weekend of aggressive "rate limits" during which even logged-in users could only read a few hundred posts per day. The company framed both moves explicitly as anti-scraping measures aimed at AI companies harvesting data.
- Early 2024 — the end of guest access. The quiet removal of "guest account" functionality — the mechanism that let unauthenticated apps fetch public content — killed most remaining third-party front ends. Nitter, the best-known privacy-friendly Twitter mirror, announced its shutdown in early 2024 for exactly this reason.
Since then, logged-out access has been in a state of permanent experimentation. Individual tweet pages usually render (they're valuable for search traffic), profile pages sometimes show a limited, non-chronological sample of posts, and everything else — search, replies, threads, trends — sits behind the wall.
Why a platform builds a wall
Three motives explain nearly everything:
- Data has a price now. When large language models made public text valuable, giving it away for free stopped making business sense. Requiring login lets the platform meter access and sell data through official (paid) channels instead.
- Logged-in users are worth more. An anonymous reader generates a page view. A logged-in reader generates a profile: interests, engagement patterns, ad targeting data, and a push-notification channel to pull them back tomorrow.
- Walls create sign-ups. A percentage of people who hit a login prompt will create an account rather than leave. From a growth team's perspective, the wall is a conversion funnel.
What was lost along the way
The costs of the closed model fall on groups that rarely show up in a quarterly report. Researchers lost the ability to study public discourse at scale — hundreds of academic projects depended on the old API. Journalists lost easy verification of public statements. Ordinary readers in low-trust situations — checking a politician's claim, following emergency information during a disaster — lost a source that didn't demand identification first. And the archival web lost coverage: pages that can't be crawled can't be preserved.
There's also a subtler cost: link rot in conversations. Millions of embedded tweets and shared links across the web now render as sign-in prompts or empty boxes for a large share of visitors, degrading pages that were written years before the wall existed. Our piece on how tweet embeds work covers why some embeds survive and others don't.
Is this the new normal?
Probably, and not just for X. Reddit restricted its API in the same year for the same reasons; other platforms have tightened logged-out access or moved content behind app-install prompts. The era in which "public social media post" meant "readable by anyone with a browser" appears to be over, replaced by a spectrum of partial visibility that platforms adjust at will. Newer networks like Mastodon and Bluesky have made openness a founding principle, which is one reason they're worth watching — we compare them in our guide to alternatives for open browsing.
For now, if you just want to read public posts without creating an account, it's still possible — with limitations and workarounds we detail in our practical guide. But it's worth being clear-eyed about the trend: the wall wasn't an accident or a bug. It's a business model, and it's working as intended.