ByteDance of TikTok fires intern for “maliciously interfering” with AI model and refutes exaggerated damage reports
TikTok’s parent company, ByteDance, has terminated an intern for “maliciously interfering” with the training of one of its artificial intelligence (AI) models. However, the tech giant has downplayed claims about the severity of the damage, calling them “exaggerated and inaccurate.”
This follows reports over the weekend that circulated widely on social media, suggesting significant disruption to ByteDance’s AI systems. The company’s Doubao AI model, a ChatGPT-like chatbot, is currently the most popular AI of its kind in China.
In a statement, ByteDance clarified, “The individual was an intern with the [advertising] technology team and has no experience with the AI Lab,” further emphasizing that “their social media profile and some media reports contain inaccuracies.”
The company also assured that its online commercial operations, including its large language AI models, were not impacted by the intern’s actions.
ByteDance denied allegations that the incident resulted in more than $10 million (£7.7 million) worth of damage by disrupting the training system, which is powered by thousands of graphics processing units (GPUs).
In addition to firing the intern in August, ByteDance reported the incident to the intern’s university and relevant industry bodies.
As one of the world’s leading social media companies, ByteDance operates TikTok and its Chinese counterpart Douyin. Known for its advanced algorithms, the company has heavily invested in AI technology to enhance applications such as Doubao, its chatbot, and Jimeng, a text-to-video tool.
