Facebook Whistleblower Alleges Meta’s AI Model Llama Was Used to Help DeepSeek
A former Facebook employee and whistleblower, Sarah Wynn-Williams, has testified that Meta’s AI model, Llama, was allegedly utilised to aid DeepSeek. During her statement to U.S. Senators, she indicated that Meta’s operations in China deserve further scrutiny, highlighting contradictions in the company’s portrayal of its activities in the country.
Key Points
- Wynn-Williams alleges Meta’s Llama AI model was integrated into DeepSeek’s operations.
- She pushed for continued investigations into Meta’s relationship with Chinese authorities concerning AI development.
- The whistleblower claims Facebook developed “custom built censorship tools” for the Chinese Communist Party.
- Concerns were raised about a “virality counter” that scrutinised popular posts for potential censorship.
- Testimony suggested that American users’ data could be potentially accessed by Chinese officials.
Why should I read this?
This article unveils serious allegations regarding Meta’s engagement with artificial intelligence in China, shedding light on the implications for data privacy and censorship practices within social media platforms. The revelations prompt reflection on the ethics of tech companies operating internationally and their accountability in addressing partnership with foreign governments.
“`