U.S. agencies say top Chinese AI companies systematically copied American models
Analysis Summary
The article reports that U.S. intelligence and cybersecurity agencies accuse major Chinese AI companies of copying American AI models like ChatGPT and Gemini through a technique called distillation, claiming this is central to China’s AI development. It highlights tensions between the U.S. and China over technology theft, with Chinese officials denying the allegations and calling for cooperation instead. The piece emphasizes the U.S. government’s view that China benefits from American innovation without creating its own.
Cross-Outlet PSYOP Detected
This article is part of a narrative being pushed across multiple outlets:
FATE Analysis
Four dimensions of psychological manipulation: how content captures Focus, exploits Authority, triggers Tribal identity, and engineers Emotion.
Focus signals
"In an alert published Tuesday, the FBI, the National Security Agency and the Cybersecurity and Infrastructure Security Agency said top Chinese AI companies have systematically mined cutting-edge AI models from American companies since 2024..."
The article opens with a time-stamped, multi-agency government alert, creating a 'breaking news' framing that signals urgency and novelty. This structure captures attention by implying a significant, coordinated intelligence disclosure.
"The sheer scale of these campaigns and their sophistication indicate that distillation is not a supplement to these companies’ AI model development, but the critical core of it"
This quote frames the activity as not just common practice but central and systemic, suggesting an unprecedented level of dependency by Chinese firms on U.S. models—elevating the perceived severity beyond routine IP concerns.
Authority signals
"In an alert published Tuesday, the FBI, the National Security Agency and the Cybersecurity and Infrastructure Security Agency said..."
The article leads with the combined authority of three major U.S. national security agencies, lending maximum institutional weight to the claims. This invocation is not merely reporting—it positions the accusation as a consensus of high-level state intelligence, which inherently pressures public acceptance.
"In April, the director of the White House Office of Science and Technology Policy, Michael Kratsios, circulated a memorandum to the heads of federal agencies highlighting Chinese distillation of American AI systems as a critical issue."
By citing a high-ranking federal science official who formally elevated the issue across government agencies, the article leverages expert bureaucratic authority to reinforce the legitimacy and urgency of the claim, going beyond neutral reporting into credibility amplification.
Tribe signals
"The U.S. has accused China for years of stealing intellectual property from American companies."
This sentence embeds the current report within a long-standing narrative of U.S.-China antagonism, reinforcing a binary framework where 'American innovation' is under siege by a foreign adversary. It converts a technical issue (AI model distillation) into a national rivalry.
"Michael Kratsios... wrote that distillation allowed Chinese companies to freeload off American innovation and circumvent safety guardrails built into American systems."
The use of 'freeload' moralizes the behavior, casting American developers as hardworking innovators and Chinese actors as parasitic. This weaponizes professional and national identity, turning AI development into a tribal loyalty test around 'fairness' and 'ownership'.
Emotion signals
"large-scale, covert industrial distillation aimed at stealing proprietary U.S. technology and undermining American research is unacceptable"
The phrase 'stealing proprietary U.S. technology' and 'unacceptable' are emotionally charged, especially when attributed to a senior U.S. official. The language escalates the act from technical copying to moral transgression, engineering outrage over perceived injustice.
"Legitimate AI distillation used to create smaller, more efficient models plays a vital role in this open innovation ecosystem"
This quote establishes a moral hierarchy: some distillation is 'legitimate' (implying U.S. practices), while other forms are not. It positions American actors as ethical participants in an 'open innovation ecosystem,' fostering a sense of moral superiority among readers aligned with U.S. tech values.
Narrative Analysis (PCP)
How the article reshapes thinking: Perception (what beliefs are targeted), Context (what information is shifted or omitted), and Permission (what behavior is being encouraged).
The article is designed to produce the belief that Chinese AI companies are systematically and covertly extracting proprietary American AI models through distillation, relying on this method as the core of their development rather than genuine innovation. This positions China’s AI advancement as parasitic on U.S. technological leadership, implying a threat to American economic and technological supremacy.
The article frames distillation as an illegitimate, large-scale operation conducted 'likely with Chinese government awareness,' which shifts the context from a competitive tech rivalry to a national security threat. This makes aggressive U.S. countermeasures—such as restricting access or escalating diplomatic tensions—feel like a necessary defense rather than an overreaction.
The article omits that distillation, while potentially violating terms of service, is a widespread and sometimes legally ambiguous practice across the global AI industry, not exclusive to Chinese firms. It also does not clarify whether the U.S. government or companies have provided verifiable forensic evidence (e.g., code-level comparisons) proving direct model copying, nor does it acknowledge U.S. companies’ own use of competitive intelligence or model benchmarking, which could normalize the behavior.
The reader is nudged toward supporting stricter U.S. controls on AI access, greater inter-agency coordination to monitor foreign usage, and a hardened stance in U.S.-China tech diplomacy—potentially including sanctions or restrictions on collaboration. Emotionally, it encourages suspicion of Chinese technological achievements and solidarity with U.S. innovation protection efforts.
SMRP Pattern
Four manipulation maintenance tactics: Socializing the idea as normal, Minimizing concerns, Rationalizing with logic, and Projecting blame.
"The spokesperson for the Chinese Ministry of Foreign Affairs said... 'rather than making groundless accusations.' This response reframes U.S. allegations as baseless, shifting blame from Chinese companies' actions to the U.S. for 'making accusations' without accepting responsibility for the alleged distillation."
Red Flags
High-severity indicators: silencing dissent, coordinated messaging, or weaponizing identity to shut down debate.
"The quote from Mao Ning — 'We always believe that all parties should strengthen cooperation to foster open, inclusive, beneficial and ethical development of AI for the well-being of all humanity' — reads as a formal, diplomatic script that avoids addressing the specific allegations and instead promotes a generalized, values-based rebuttal, consistent with state messaging rather than spontaneous response."
Techniques Found(4)
Specific propaganda techniques identified using the SemEval-2023 academic taxonomy of 23 techniques across 6 categories.
"large-scale, covert industrial distillation aimed at stealing proprietary U.S. technology and undermining American research is unacceptable"
Uses emotionally charged language ('covert,' 'stealing,' 'undermining') to frame the practice of distillation as inherently malicious and unethical, going beyond the neutral or technical description of the act and implying moral condemnation without presenting evidence of intent or illegality beyond alleged violations of terms of service.
"the FBI, the National Security Agency and the Cybersecurity and Infrastructure Security Agency said"
Cites multiple high-authority U.S. government agencies to lend credibility to the claim about Chinese AI companies' actions, potentially discouraging scrutiny by positioning the assertion as an official consensus rather than a contested allegation.
"The sheer scale of these campaigns and their sophistication indicate that distillation is not a supplement to these companies’ AI model development, but the critical core of it"
Exaggerates the centrality of distillation in Chinese AI development by asserting it is the 'critical core' rather than one component among many, based on circumstantial evidence and without providing comparative analysis of other development methods used by the companies.
"freeload off American innovation"
Uses a pejorative and economically charged term ('freeload') to characterize Chinese companies' use of distillation, implying parasitic behavior and moral irresponsibility without engaging with whether such practices are common or accepted in the broader AI industry.