GPT-4o: Meta regularly briefed members of the Chinese Communist Party on artificial intelligence, including facial recognition, photo tagging, and Facebook’s Llama AI model​...

Question 1 of 2 for ChatGPT-4o: Please analyze the transcript of the full hearing of the Facebook whistleblower before the Senate Judiciary Committee and list all surprising, controversial and valuable statements made, for citizens, regulators and businesses that care about fair competition, data privacy and national security.

ChatGPT-4o’s answer:
Based on the transcript of the full Senate Judiciary Committee hearing with Facebook whistleblower Sarah Win Williams, here are the most surprising, controversial, and valuable statements relevant to citizens, regulators, and businesses concerned with fair competition, data privacy, and national security:

🔥 SURPRISING Statements

1. Facebook built censorship tools for China:

   • Meta (Facebook) designed and tested censorship tools explicitly for the Chinese Communist Party (CCP), including a “Chief Editor” function that could block viral content and even shut down services in entire regions like Xinjiang or on politically sensitive anniversaries like Tiananmen Square​.

2. China access despite public denials:

   • Meta secretly conducted business in China since at least 2014, despite public claims that they had no operations there. Their own SEC filings show China is Meta’s second largest market, primarily through advertising revenue and business partnerships​.

3. Meta shared AI advances with CCP:

   • Meta briefed the CCP on artificial intelligence, including facial recognition and photo tagging, which may have contributed to China's military and surveillance technology, such as the development of the DeepSeek AI model​.

4. $50,000 per mention gag order:

   • Meta imposed an extreme gag order on Sarah Win Williams, threatening $50,000 in fines each time she publicly mentioned Facebook, even if her statements were true​.

5. Zuckerberg’s personal investment in China:

   • Mark Zuckerberg learned Mandarin, traveled to China more than any other country, and personally led Project Uldren, Meta’s secret mission to enter the Chinese market​.

🧨 CONTROVERSIAL Statements

1. Deliberate censorship of a Chinese dissident on U.S. soil:

   • Meta removed the Facebook account of a Chinese dissident living in the U.S. at the request of the CCP—then lied about it under oath to Congress​.

2. Willingness to compromise U.S. user data:

   • Facebook proposed building POP servers on Chinese soil, where U.S. user data would be mixed with Chinese data, exposing Americans to surveillance by the CCP​.

3. Meta helped China bypass American competitors:

   • Meta pitched itself to China as a way to avoid reliance on Cisco and IBM, directly undercutting fellow American tech firms​.

4. Internal dissent ignored:

   • Facebook security engineers raised strong objections to China plans, but were overruled. One said: “My red line is not Mark Zuckerberg’s red line.”​

5. Executives shielded their own children:

   • Meta’s executives didn’t allow their own teenagers to use Facebook or Instagram, acknowledging harms even as they denied them publicly​.

💡 VALUABLE for Policy & Business Strategy

1. Modeling Big Tech after Big Tobacco:

   • Meta’s board used Big Tobacco’s PR playbook to shape narratives, including dismissing health and safety concerns while targeting children​.

2. Intentional targeting of emotionally vulnerable teens:

   • Meta knowingly targeted depressed teenagers with ads, sharing data with advertisers based on signs like deleted selfies or phrases like “I feel worthless”​.

3. Use of AI model Llama in Chinese military applications:

   • Meta's open-source Llama AI model has been used in China’s AI military development and in products like DeepSeek. Open-sourcing it may strategically benefit Meta by raising its importance in a global “AI arms race”​.

4. Legal manipulation to silence whistleblowers:

   • Meta reversed its public 2018 commitment to forgo forced arbitration with whistleblowers and instead pursued Win Williams for hundreds of millions of dollars​.

5. Meta’s misrepresentations to Congress:

   • Multiple senators highlighted how Meta lied repeatedly to Congress, including about user data practices, censorship tools, and China dealings​.

Question 2 of 2 for ChatGPT-4o: List everything that is being said about Artificial Intelligence.

ChatGPT-4o’s answer:

🔍 Key Testimonies and Allegations on Artificial Intelligence

1. Briefings to Chinese Officials

• Meta (formerly Facebook) regularly briefed members of the Chinese Communist Party (CCP) on artificial intelligence, including facial recognition, photo tagging, and Facebook’s Llama AI model​​​.

• These briefings were described as part of Meta’s “value proposition” to help China outcompete U.S. companies and avoid relying on firms like Cisco and IBM​​.

2. Support for China’s AI Military Program

• Facebook’s open-source AI model Llama has been used in China’s AI military development, including in the creation of DeepSeek, a prominent Chinese large language model (LLM)​​​.

• It was acknowledged that Llama’s open-source nature, and Meta's strategy of promoting it, may have strategically benefited Meta by increasing its relevance in the “winner-takes-all” AI race​.

3. Facial Recognition and Surveillance Technology

• Facebook allegedly shared sensitive AI technologies (including facial recognition and surveillance capabilities) with Chinese security officials, enabling surveillance not only within China but also potentially outside, such as in Taiwan and Hong Kong​​​.

4. National Security Implications

• There were repeated warnings from Meta’s own engineers about the national security risks of sharing AI technology and hosting data in China. Their objections were overridden by leadership​​.

• The company contemplated storing U.S. user data in China, which would have made it accessible to the CCP, raising red flags about AI-enabled surveillance capabilities​​.

5. Strategic AI Positioning via Open Source

• Meta’s open-source Llama model was framed as a counterbalance to China’s AI advancements, allowing the company to argue that “America must win the AI race” by supporting U.S.-backed open-source alternatives​.

• This strategy could allow Meta to benefit regardless of who actually wins, since Meta positions itself as a key player in both Western and Eastern AI ecosystems​.

📌 Other Notable Mentions

• AI Weaponization: It was noted that Chinese researchers used Llama in AI weapons development, which was presented as a direct consequence of Meta’s open-source approach and technology transfers​​.

• Moonshot Projects: Meta undertook “moonshot” AI initiatives in China, but whistleblower Williams preferred to provide more details on these sensitive efforts in private due to their potential implications​.

• Calls for Oversight: The committee and senators explicitly framed this AI collaboration with China as a historic failure of corporate governance and a risk to national sovereignty, urging investigations into how and why these decisions were made at Meta​​.