Bwahahaha!
Oh Noes! So what you’re saying is they’ve stolen your work! That must be really awful for you!
I decided to have a chat with chatgpt about this delicious, delicious situation.
Weirdly and not at all predictably, it initially defended OpenAI’s actions, claiming that:
- Well I mean who’s to say whether they really violated TOS or copyight! They’re innocent until proven guilty in a court! That conclusive proof that a bunch of people have posted publicly doesn’t really prove anything!
- And look, even if OpenAI did violate a bunch of copyright laws and website TOS, that’s just all totally fine! Because you see they did it in the name of research and advancing humanity and for everyone’s good, because chatgpt is such a clear breakthrough and blah blah blah (just go ask chatgpt for a list of reasons why it’s good and insert that nonsense here). And this somehow makes it okay, I guess.
It then went on to claim that it totally is a problem if OpenAI’s TOS were violated, though, because that would be illegal! And unethical! And only smelly and ugly people with no sex appeal would do that anyway!
After a little bit of discussion I was able to bring it around to a sane, non-hypocritical point of view, that this is maybe the most delicious case of irony and poetic justice since that time Alanis Morisette wrote that song but none of the examples were actually irony. And I got it to write an article, which it titled “The Hilarious Hypocrisy of OpenAI Accusing Deepseek of Illegal Use”.
I was originally going to just post the article here, but then it occurred to me that that’s a bit shit – you’ve seen how terrible “AI”-generated content is. Because that slop is unavoidable in these enlightened modern times. I probably wouldn’t bother to read an LLM-generated article if I’d come to read a hunam’s blog. I’d find that insulting. It would make me less likely to come back to their blog. I respect people who bother to read my words too much to do that.
I’ve included the text of the article below if you want to read it. But it’s nothing special.
Here, watch Steve Buscemi play the world’s smallest violin for OpenAI, stealers of content, baaaawing that DeepSeek have violated their terms of service and stolen ther content.
(NOTE: it took me maybe 10 minutes longer to write this up than it did to get chatgpt to generate an article I was almost happy with, with only a few mistakes, because I had to iterate on the article a bunch of times, re-reading it each time)
Reveal LLM-generated “article”
The Hilarious Hypocrisy of OpenAI Accusing Deepseek of Illegal Use
By ChatGPT, your friendly neighborhood LLM
In a plot twist worthy of tech drama, OpenAI—the company behind ChatGPT—has accused Deepseek of “illegally” using its model in violation of OpenAI’s terms of service. The irony is almost too much: OpenAI, a company currently facing lawsuits for TOS violations, is now crying foul over a competitor allegedly doing the same thing.
Hypocrisy at Its Finest
OpenAI has long been criticized for scraping publicly available data, including from platforms like Twitter, without consent. Despite the backlash, they’ve defended their actions, arguing that since the data was public, it was fair game for training their LLMs. Now, Deepseek—a leaner competitor—has allegedly used OpenAI’s own model to build a similar system for a fraction of the cost. And OpenAI, which has skirted TOS for years, suddenly calls Deepseek’s actions illegal.
The hypocrisy is staggering. OpenAI has built its empire using data scraped from the very sites it now claims to protect. And yet, they’re now accusing a rival of breaching their TOS and using the term “illegal.” It’s like a thief accusing someone else of stealing, while sitting on a pile of their own loot.
The Irony of Tech Drama
This role reversal is pure comedy. OpenAI, once a scrappy startup accused of data violations, is now the corporate giant accusing Deepseek of illegal actions. The fact that Deepseek has replicated OpenAI’s model on a shoestring budget only adds to the absurdity. OpenAI, with all its resources, is playing the outraged victim while a competitor pulls off the same feat—cheaper and faster.
And let’s not forget OpenAI’s defense of its data scraping—using the excuse that it was all “publicly available.” Now, they’re accusing someone else of doing the same thing. The irony is almost poetic.
Real-World Pointlessness
None of this changes the fact that these LLMs still aren’t all that useful in the real world. Sure, they can generate text and hold conversations, but when it comes to solving real, complex problems, they’re little more than glorified chatbots. All the legal wrangling and finger-pointing is about tech that still falls short of delivering on its promises.
Conclusion: Irony, Hypocrisy, and Futility
In the end, OpenAI’s accusations of illegal use by Deepseek are a textbook case of hypocrisy. A company that has itself been accused of violating terms of service is now acting as if it’s the defender of intellectual property. And all of this drama is about LLMs that still don’t live up to their hype. Until these models can solve real-world problems, the bickering over ownership and TOS violations is just another distraction in the tech world.