Pablo Gonchar | Lightrocket | Getty Images
Twitter CEO Elon Musk threatened to sue Microsoft on Wednesday, accusing the company of illegally using data from social media companies to train artificial intelligence models.
The Musk tweet threat came after Mashable and other publications reported that Microsoft would remove Twitter from its advertising platform. This allows ad buyers to manage all their social media accounts in one place.
“They used Twitter data to train illegally,” Musk tweeted. “Lawsuit Time”
Musk, who is also the CEO of Tesla and SpaceX, has often tweeted about plans that never materialized, and apparently no lawsuits have been filed. A Twitter press line did not respond meaningfully to a request for comment, and a Microsoft representative declined to comment.
The Musk threat is the latest sign that data ownership is rapidly becoming a fierce battlefield in the generative AI rush. Big tech companies are working to develop cutting-edge AI models like OpenAI’s GPT, and data owners are trying to stop them or charge for using their content.
Microsoft has developed its own so-called Large Language Models (LLMs) and sells access to OpenAI’s models. Microsoft invested $10 billion in OpenAI last year in an unusually structured deal. Musk said he was the co-founder of OpenAI before he left the board in 2018, and recently spoke about the company’s transition from a non-profit model to a highly valuable Microsoft-influenced business. complaining.
LLMs like GPT require terabytes of data for training, much of it collected from websites such as Reddit, StackOverflow, and Twitter. Training data from social networks is valuable because it captures informal interactions.
As these new AI models move from labs and universities into the corporate world, data owners are beginning to make demands.
For example, Reddit announced earlier this week that it would charge companies for access to the programming interface used to feed conversations between Reddit users into its AI training software. Universal Music Group this week responded to a viral video of a song that it claimed used AI to imitate rapper Drake, saying such training of an artist’s music is “a breach of contract and a violation of copyright law.” said.
And stock photo database Getty Images is suing Stable Diffusion for copying its content to train an AI image generator.
Musk said in December that Twitter would “suspend” access to OpenAI’s database. He also announced plans to build his own large-scale language model in one of his companies called TruthGPT.
clock: Elon Musk wanted to take charge of OpenAI in 2018
