Tag
Executives from OpenAI and Microsoft admitted in court filings that large language models are built on stolen content and are destroying the web, creating a 'doom loop' that threatens the economic foundations of content creators.
Court documents reveal that OpenAI and Microsoft were aware their data scraping practices for AI training could create a 'doom loop' damaging the web, and constitute 'the largest theft of labor in human history,' as alleged in the New York Times lawsuit.
European lawmakers are calling for an investigation into Meta after its copyright system was allegedly weaponized to suppress accounts posting about anti-government protests in Albania, raising concerns about censorship and compliance with the EU's Digital Services Act.
Internal documents from Microsoft and OpenAI reveal executives' concerns that AI scraping of news content is the 'largest theft of labor in human history' and could create a 'doom loop' harming news organizations and model performance.
New unredacted filings in a lawsuit reveal that a Microsoft executive described AI scraping as 'the largest theft of labor in human history,' with internal admissions about AI's impact on publishers like The New York Times.
The article critiques AI detectors like Pangram and GPTZero for their low accuracy against fine-tuned AI models and references a study indicating readers prefer AI outputs trained on copyrighted books.
A Twitter user warns that friends are receiving fake DMCA takedown notices and having their handles changed, with direct messages sent as part of the scam.
A talk by Cory Doctorow discussing the history of DRM and copyright wars, and their implications for the future of general-purpose computers and society.
Steven Johnson of Google Labs explains how AI can clone a writer's voice without permission, highlighting challenges for copyright protection and the freelance economy.
Anime reaction YouTubers are facing a wave of false copyright strikes from a company called Remove Your Media, which represents clients like Crunchyroll and Viz Media, leading to temporary video takedowns and concerns over fair use.
The author criticizes the idea that AI models stole solutions to the millennium problems from scientists' training data, arguing that researchers were not close to solving these issues independently.
Suno has launched Suno v6, a new AI music model family trained on licensed data from major music labels, aiming to address copyright lawsuits and introduce enhanced features for music creation.
This article reports that two mathematicians claim OpenAI used their private work to train its Sol and Astra models before they could publish their own findings.
A tweet suggesting the urgent reverse engineering of closed source software to preempt copyright-based restrictions.
Mother Jones is suing OpenAI for using their stories without permission to train ChatGPT, highlighting issues of copyright infringement and fair use in AI training data.
The US government has filed a brief supporting OpenAI in a copyright lawsuit, arguing that restricting the use of copyrighted material for training LLMs could hinder AI innovation and US leadership in the field.
Anthropic updated Claude's system prompt to include strict restrictions on reproducing song lyrics, likely in response to copyright lawsuits, and the article details these changes and their implications.
The United States has filed a legal statement in OpenAI copyright lawsuits, arguing that AI training on copyrighted works is fair use and criticizing the 'market dilution' theory as deeply flawed.
Google is pursuing licensing agreements with Hollywood studios to train its AI models on copyrighted content, aiming to enhance public perception of AI while potentially posing risks to the studios involved.
Music publishers including Sony are suing Anthropic for allegedly pirating copyrighted songs to train its AI, with internal chats extolling piracy cited as evidence.