Member-only story
OpenAI vs. DeepSeek R1: The Irony of AI Copyright Battles
The recent battle between OpenAI and DeepSeek R1 has been nothing short of entertaining. OpenAI is adamant that DeepSeek R1’s distillation of its models for commercial use is illegal. According to OpenAI, their terms of service strictly prohibit this, making R1’s actions an outright violation. However, when you take a step back, the situation becomes laughably ironic.
The Pot Calling the Kettle Black?
OpenAI’s complaints about unauthorized usage seem rather hypocritical when you consider its history. Let’s not forget that OpenAI has been accused of training its models on vast amounts of GPL-licensed code from GitHub. Despite the likely breaches of hundreds of software licenses, OpenAI has not reciprocated by making its models open-source.
Even more eyebrow-raising is the suggestion that OpenAI has trained its models on copyrighted images and social media posts without explicit permission. Twitter, for example, allows posts to be available for its platform, but at no point did it grant OpenAI or any third party the right to scrape and utilize them for AI training. And yet, OpenAI now claims to be the victim of unauthorized usage. The irony is staggering.
