A new Red Scare is sweeping through Silicon Valley’s AI industry, as Chinese-born professionals face mounting suspicion and accusations of espionage amid escalating US-China tensions. The issue came to a head at an AI conference after-party in Berkeley last November, when a 27-year-old Chinese woman named Sarah—who had left her trading career to work in AI safety—was publicly accused of being a spy on social media after a casual conversation about Chinese AI policy.
The incident sparked millions of views on X (formerly Twitter), with Samuel Hammond, an AI policy advisor for Project 2025, posting about the encounter and speculating about Sarah’s intentions. The viral post forced Sarah to stop attending networking events and ultimately return to China in December, abandoning her AI safety career aspirations in the United States.
This atmosphere of suspicion has intensified as American tech companies race to develop advanced AI systems before China. The fear of Chinese espionage—the same concern that led to TikTok’s shutdown—is now pushing out Chinese-born AI professionals during a critical talent shortage in the field. According to the Financial Times, Google, OpenAI, and several other US tech companies have tightened personnel screening, with some startups using third-party tools like Strider Technologies to investigate employees’ connections to “state-sponsored risk.”
The current climate builds on years of escalating tensions. Trump’s controversial “China Initiative” launched in 2018 investigated hundreds of academics suspected of ties to China but found only a few cases of actual spying. Most cases involved undisclosed funding rather than espionage. Biden’s administration continued the competitive approach with the 2022 CHIPS Act, restricting US investments and research collaborations in China.
Influential figures in Silicon Valley have amplified these concerns. AI researcher Leopold Aschenbrenner published a widely-circulated manifesto warning about an “all-out race with the CCP” for artificial general intelligence (AGI), which he called “the most powerful weapon mankind has ever created.” Venture capitalist Marc Andreessen suggested that “all such American AI labs are fully penetrated” by China.
The impact on Chinese diaspora workers is severe. Many report being called “honeypots” or spies in social situations, turning down AI job opportunities to avoid scrutiny, and downplaying their cultural identity. Chinese startup founders are removing references to China from pitch decks and refusing Chinese investment capital. HeyGen, a generative-AI startup, dissolved its Chinese operations entirely to relocate to the US. Visa processing delays are stranding some workers abroad for months or years, with one AI Ph.D. student waiting over 130 days to return.
Key Quotes
I’m a Chinese national, but it’s not like I’m a spy
Sarah made this comment to another attendee at an AI conference after-party, trying to clarify an uncomfortable conversation. The remark was overheard by Samuel Hammond, who posted it to X where it went viral with millions of views, ultimately derailing her AI safety career and forcing her to leave the United States.
Whenever you say anything neutral about China, people will think, ‘That’s pro-China, and this person is bought by the CCP.’
Sarah explained how the hawkish environment in Silicon Valley makes it nearly impossible for Chinese nationals to engage in nuanced discussions about AI policy or bridge the information gap between Chinese and English-language researchers—the very work she hoped to contribute to AI safety efforts.
If we’re lucky, we’ll be in an all-out race with the CCP; if we’re unlucky, an all-out war
Leopold Aschenbrenner wrote this in his widely-circulated manifesto ‘Situational Awareness: The Decade Ahead,’ which compared AGI to nuclear weapons and painted a dystopian scenario of Chinese espionage. The paper, shared by Ivanka Trump, has significantly influenced Silicon Valley’s perception of Chinese involvement in AI.
We work for no government. We just want to build businesses.
A Chinese founder told journalist Rita Liao this statement, capturing the frustration of Chinese entrepreneurs caught between two governments’ suspicions. Many are now refusing Chinese investment, removing references to China from materials, and relocating operations to avoid scrutiny from both US and Chinese authorities.
Our Take
This situation represents a troubling paradox in America’s AI strategy. The tech industry’s response to espionage concerns—heightened screening, AI-powered surveillance of employees, and social ostracism—mirrors the authoritarian tactics it claims to oppose. The Trump-era China Initiative’s track record should serve as a cautionary tale: it destroyed careers over bureaucratic oversights while catching few actual spies. Yet Silicon Valley appears to be repeating these mistakes with even less oversight. The real casualty may be America’s competitive advantage itself. AI development thrives on diverse perspectives and open collaboration—the very things this new Red Scare is eliminating. When talented researchers like Sarah flee the US, and Chinese founders restructure their companies to appear less Chinese, we’re witnessing a brain drain driven by xenophobia rather than security. The question isn’t whether espionage concerns are legitimate—they are—but whether the current approach serves American interests or undermines them. A more sophisticated strategy would distinguish between genuine security threats and valuable contributors, but nuance is the first victim of geopolitical panic.
Why This Matters
This story reveals a critical tension at the heart of America’s AI competitiveness strategy. While the US government and tech industry emphasize the urgency of winning the AI race against China, the atmosphere of suspicion is simultaneously driving away talented Chinese-born AI researchers and engineers during an acknowledged talent shortage. The irony is stark: efforts to protect American AI leadership may be undermining it by excluding a significant pool of qualified professionals.
The implications extend beyond individual careers to the broader AI ecosystem. Chinese founders are restructuring their companies, refusing Chinese investment, and relocating operations to avoid scrutiny—decisions that reshape the global AI landscape. The chilling effect on international collaboration could slow AI safety research, which Sarah and others see as crucial for humanity’s future. As geopolitical tensions intensify and AI capabilities advance, this dynamic creates a “double bind” where Chinese nationals face restrictions from both governments, potentially fragmenting the global AI research community at a time when cooperation on AI safety and ethics may be most needed. The situation also raises questions about how democracies balance legitimate security concerns with the openness and diversity that have historically driven technological innovation.
Recommended Reading
For those interested in learning more about artificial intelligence, machine learning, and effective AI communication, here are some excellent resources: