censorship

A research project demonstrates that distilling a heavily censored Chinese AI model (DeepSeek) into an American model (GPT-OSS) does not transfer the censorship behavior, while performance gains are retained, specifically in financial reasoning.

India ordered GitHub to remove the Bluetooth messaging app BitChat within three hours, but the codebase was quickly mirrored on the decentralized platform Radicle, making removal impossible and demonstrating the power of decentralized infrastructure.

Jack Dorsey tweets that the Indian government dislikes technologies like Bitchat and wants them taken down.

Google is removing user reviews from Google Maps for ICE detention centers, erasing accounts of experiences and protests, citing policy against reviews for places people go without choice. Research by Tulane professor Muira McCammon documents the suppression.

Discusses the technical method of triangulation by testing network latency to identify and block Chinese users from using AI services (like Claude).

A Reddit user was permanently banned from a subreddit for accusing an AI-generated account of being fake, highlighting concerns about automated moderation silencing dissent.

The article exposes that Anthropic secretly detects and marks Chinese users in Claude Code using steganography (modifying Unicode characters and separators in date strings) for account bans, sparking strong community concerns about privacy and trust.

A tweet from Ahmad Osman quotes Harrison Kinsley expressing concern that the public's fear of AI and job loss will lead to severe restrictions on open source AI and potential imprisonment, urging people to take the issue more seriously.

A global movement opposing government restrictions on the open internet, launching a website to track censorship efforts and partner with digital rights organizations.

A blog post examines the coordinated global push for internet surveillance and censorship, and laments that public will to fight for internet freedoms has waned as the internet has become more centralized, contrasting with past mass movements like the SOPA protests.

The Pragmatic Engineer blog reports that Pollen's CEO and CTO are using bogus copyright claims to remove a critical article about the company's collapse from Google search results, highlighting abuse of the DMCA takedown system.

Cory Doctorow reports on Mark Zuckerberg's escalating legal war against whistleblowers, including a demand for $111 million and a gag order on a book author.

The article reports that major Hollywood studios have declined to distribute Luca Guadagnino's film 'Artificial' about Sam Altman's ouster from OpenAI, following Amazon MGM's withdrawal, raising concerns about Big Tech's influence on critical storytelling.

Internal documents reveal that the Israeli government repeatedly requested Meta to censor posts and accounts about the Iran war, and Meta complied with some requests without disclosing how many.

Noam Shazeer's departure from Google to OpenAI is reportedly due to Google management censoring his comments on transphobic and Gaza conflict topics.

Salvatore Sanfilippo speculates that OpenAI might delay GPT-6 (or 5.6) to avoid being blocked like the Fable incident, suggesting they could selectively publish benchmarks and release a censored model for cybersecurity.

Argues that because LLMs must encode harmful content to identify it and jailbreaks are always statistically possible given large user bases, there is a non-zero chance of harm; the author therefore advocates against censorship to ensure good actors have the same tools as bad actors.

The White House is negotiating with Congress to block state-level AI regulation in exchange for advancing three federal internet censorship bills, drawing bipartisan backlash.

Elon Musk criticizes a proposed UK censorship law, arguing it is a wolf in sheep's clothing that would enable the government to track everyone, while quoting a user who claims the law effectively bans social media for everyone unless they identify themselves.

The Trump administration ordered Anthropic to shut down its Fable 5 AI model using export law, marking the first time federal regulators have removed a deployed AI system from public access.