Keywords are those words that frequently occur in the target corpus assessed by keyness values. A higher keyness value represents a more significant distribution of its corresponding keyword. Table 2 demonstrates the top 25 keywords in the news reports on AI in CD and NYT. These keywords are calculated by Wmatrix 7 and ranked according to their keyness values (i.e. log-likelihood values) with BNC 1994 Sampler Written as the reference corpus.

Table2. Top 25 Keywords in the news report on AI in CD and NYT.

Figure1. The hierarchical cluster of top 25 keywords in CD.

Figure2. The Hierarchical cluster of Top 25 keywords in NYT.

Five clusters in CD shown in Figure 1Figure 1 are centered around technological innovations, industrial applications, and global governance of AI, as well as digital transformations driven by AI. Cluster 1 includes “Alibaba” and “cloud”. Alibaba Cloud is a representative cloud computing platform affiliated with Alibaba Group, offering a broad range of services such as big data analytics and database management to serve the development of AI infrastructures in China. Cluster 2 includes “Chinese”, “Baidu” and “AI-powered”, focusing on the practices of Chinese localized enterprises in AI innovations and applications. Baidu, as a major Chinese internet and AI company, endeavors to integrate AI technologies into products and services, thus facilitating the implementation of AI-powered strategies. Cluster 3 includes “AI”, “technology”, “China”, “development”, “innovation”, “industrial”, “industry” and “application”. This cluster is the one that is most directly associated with AI development, indicating CD’s considerable emphasis on the innovations and applications of AI across various sectors. Cluster 4 includes “global” and “governance”, encompassing topics such as international cooperation and the establishment of a global governance system for AI. CD situates AI within an international context, underscoring the shared responsibilities undertaken by different countries in building responsible AI. Cluster 5 includes “data”, “llms”, “digital”, “intelligent”, “education” and “Shanghai”, representing the digital transformation driven by AI. Among these keywords in Cluster 5, “education” especially stands out and is represented along two dimensions specifically including the cultivation of AI talents and the capacities of AI to empower educational practices.

Five clusters in NYT shown in Figure 2Figure 2 present multifaceted topics encompassing technological actors, GenAI applications, digital infrastructures, political communication, and technological competitions. Cluster 1 includes “Microsoft”, “Google”, and “Meta”, indicating NYT’s strong focus on corporate actors in AI-related news coverage. Cluster 2 includes “generate”, “OpenAI”, and “ChatGPT”, referring to the breakthroughs of new-generation AI technologies exemplified by ChatGPT. ChatGPT released by OpenAI serves as a milestone in GenAI development, exerting profound impacts on industrial ecosystems and social interactions. Cluster 3 includes “Internet”, “digital”, “data”, and “researchers”, highlighting digital environments, data resources, and research communities that enable AI innovations. Cluster 4 includes “fake”, “Biden”, and “Trump”, presenting the intersections between AI and political communication. NYT emphasizes that AI-generated content may intensify uncertainties within political communication in that the misinformation and disinformation brought by AI may manipulate public judgment and pose challenges to the reliability of electoral information during presidential campaigns. Cluster 5 includes “U.S.”, “AI”, “technology”, “Musk”, “Nvidia”, and “executive”. This cluster mainly denotes the prominent entrepreneurs, technological companies, and corporate decision-makers in the landscape of AI industry, thereby suggesting the pivotal role of AI in shaping future technological competitions.

The paragraphs positioned by the top 25 keywords in CD and NYT are further extracted to facilitate the identification of conceptual metaphors. These extracted paragraphs specifically comprise 39,436 tokens in CD (91.37% of the total), and 40,653 tokens in NYT (88.74% of the total). Next, following the identification procedures introduced in section 4, the distributions of conceptual metaphors identified across CD and NYT are shown in Table 3 below.

Table3. Distributions of conceptual metaphors identified across CD and NYT.

Note: a. * means p ≤ 0.05; ** means p ≤ 0.01; *** means p ≤ 0.001.

Table 3 indicates that both CD and NYT employ a wide range of conceptual metaphors, whose overall distributions do not exhibit significant differences (x² = 0.24, p = 0.63 > 0.05) on the whole. 870 metaphorical representations are identified in CD, with the metaphor type of BUILDINGS AND CONSTRUCTION being the most frequent (Freq. = 184, Pro. =2 1.15%), followed by FORCES (Freq. = 139, Pro. = 15.98%) and TRAVEL (Freq. = 112, Pro. = 12.87%). This distribution suggests that CD prefers to construe AI as an entity and process that can be built, propelled, and progressively developed. By contrast, NYT records 875 metaphorical representations, with the metaphor type of TRAVEL ranking first (Freq. = 173, Pro. = 19.77%), followed by BUILDINGS AND CONSTRUCTION (Freq. = 147; Pro. = 16.80%) and WAR (Freq. = 98; Pro. = 11.20%). It can be seen that in addition to development narratives, NYT puts a stronger emphasis on the potential competitions and conflicts derived from AI-related issues than CD. Chi-square tests further unravel significant cross-media differences in several metaphor types. To be more specific, CD employs PLANTS (x² = 7.80, p = 0.005 < 0.01), BUILDINGS AND CONSTRUCTION (x² = 6.43, p = 0.01), FORCES (x² = 14.40, p = 0.001), and PERFORMANCE (x² = 5.25, p = 0.02 < 0.05) metaphors significantly more frequently than their counterparts in NYT. NYT significantly favors ANIMALS (x² = 4.83, p = 0.03 < 0.05), GAMES AND SPORT (x² = 8.77, p = 0.003 < 0.01), WAR (x² = 7.72, p = 0.005 < 0.01), and TRAVEL (x² = 10.52, p = 0.001) metaphors.

Obviously, the similarities and differences can be found in the distributions of conceptual metaphors identified in CD and NYT. Given the limited space, this sub-section mainly narrows down the focus to the top three types of conceptual metaphors respectively in CD and NYT. Besides, the other four types of conceptual metaphors with significant differences between CD and NYT are considered as well.

5.3.1. Representative Conceptual Metaphors in News Reports on AI by ChinaDaily

The three most frequently distributed metaphor types in CD are BUILDINGS AND CONSTRUCTION, FORCES, and TRAVEL, collectively conceptualizing AI development as a continuous process of construction, empowerment, and forward advancement.

(1)BUILDINGSANDCONSTRUCTIONMetaphor

BUILDINGS AND CONSTRUCTION metaphor in CD is closely connected to AI development. A series of structural features represented by tokens such as “build*”, “establish*”, and “construct*” demonstrates that AI IS A BUILDING or AI IS A FOUNDATION. On the one hand, AI can be deemed as a sociotechnical system that requires multi-stakeholder participation and continuous investment. Relevant reports in CD call for building a favorable technological environment, an innovative AI talent ecosystem, and an inclusive AI global governance system, etc. In Example (1), AI IS A BUILDING, the governance of which is supported by the collective efforts of government, companies and societies. On the other hand, AI development underpins the attainment of broader technological goals such as “building a Digital China”. Discursive representations exemplified by Example (2) imply AI IS A FOUNDATION, thus reinforcing the strategic significance of AI in the national narratives at the macro level.

Example (1): It is good that the discussions focused on actions to synergize the efforts of governments, companies and societies to build an inclusive AI global governance system to ensure that the technology serves the common good of humanity.

(CD, February 11, 2025)

Example (2): According to an action plan released by the National Data Administration in May, China aims to make great strides in building a Digital China by the end of 2025, with the continuous expansion of new quality productive forces and improvements in the quality and efficiency of economic development.

(CD, July 26, 2025)

(2)FORCESMetaphor

FORCES metaphor in CD is typically represented by “enhance*”, “boost*”, “breakthrough”, “push” and “disruptive”, etc., putting AI into a dynamic network featured by the incessant interplay of forces. Further to this, two main types of FORCES metaphors have been identified in CD, specifically including AI IS AN EXERTING FORCE and AI IS A RECEIVING ENTITY. AI itself as a disruptive power exerts both positive and negative influences on the whole society. A vast majority of relevant concordances in CD emphasize the benefits brought by AI particularly by means of its industry-specific applications. For instance, five dynamic force-related tokens underlined in Example (3) together conceptualize AI as a force propelling industrial transformation and high-quality economic development.

Example (3): Policies encourage leveraging AI to empower industrial production and accelerating the upgrading of high-end equipment, which help to advance the high-quality development of the industrial internet and smart manufacturing, enhance the core competitiveness of the manufacturing sector and drive the modernization of industrial and supply chains, Zheng said.

(CD, August 30, 2025)

Concordances indicating that AI IS A RECEIVING ENTITY primarily highlight how external forces facilitate AI development. For example, universities are supportive forces that strengthen AI talent cultivation; international cooperation enhances the precision of AI regulation and AI governance; and national initiatives advance the development of AI governance system.

(3)TRAVELMetaphor

TRAVEL metaphor in CD emphasizes AI development as a continuous process with a clearly defined direction, a discernible journey, and an accelerating pace, thus surrounded by a direction scenario, a path scenario, and a pace scenario. The direction scenario mainly focuses on technological innovation and ethical governance of AI. In terms of technological innovation, multimodal GenAI is positioned as the direction of GenAI development to pursue artificial general intelligence (Example 4). In terms of ethical governance, CD puts more emphasis on directions aligning with international cooperation (Example 5).

Example (4): Multimodal large language models have made substantial advances over the past year, and the practical application of such models is heading in the direction of pursuing artificial general intelligence, with diverse vertical industrial large models and AI agents emerging, said experts at the 2024 World AI Conference, which wrapped up on July 6 in Shanghai.

(CD, July 17, 2024)

Example (5): The themes of the Paris summit, including public interest AI, the future of work, innovation and culture, trust in AI and global AI governance, highlight some of the contentious issues related to the technology, as well as the directions in which collective efforts are called for to address the challenges.

(CD, February 11, 2025)

The path scenario of TRAVEL metaphor in CD is typically represented by “path*” and “journey”. Relevant concordances depict the transitions of AI from isolated breakthroughs to a mature AI ecosystem, the endeavors made by China to boost AI governance, and the innovations led by technological companies, etc. However, of paramount emphasis is still the integration of AI with specific industries, which resonates with “AI Plus” strategies advocated in China. Taking the education field as an example, Example (6) spatializes the abstract process of educational innovation by conceptualizing the applications of AI as “new paths” to facilitate enhanced learning outcomes. The pace scenario of TRAVEL metaphor in CD is represented by “stride*” and “pace*”, etc., which is employed to highlight the rapid advancement of AI and its expanding societal impacts. For instance, Chinese data sector is reported to have made “great strides” in Example (7), making a solid foundation for data-related AI development.

Example (6): Liu Jun, rector of the City University of Macao, said the application of AI in English education has opened up new paths for personalized learning, innovative teaching methods and language proficiency enhancement.

(CD, July 27, 2024)

Example (7): China’s data sector has made great strides during the 14th Five-Year Plan (2021-25) period.

(CD, August 30, 2025)

5.3.2. Representative Conceptual Metaphors in News Reports on AI by TheNewYorkTimes

Three most frequently distributed metaphor types in NYT are TRAVEL, BUILDINGS AND CONSTRUCTION, and WAR, indicating that AI development is a process of progress, construction, and competition.

(1)TRAVELMetaphor

TRAVEL metaphor with the most frequent distribution in NYT can be perceived by “advance*”, “goal*” and “reach*” etc. These representations primarily evoke scenarios of forward movement and destination. Interestingly, most of the concordances representing seemingly positive forward movement scenarios possess a negative tone in NYT, criticizing that the advancement of AI may place society beyond human control and disrupt the political sphere through the widespread democratization of disinformation, etc. Particular negative attention has also been paid to the advancement of AI in China. For instance, Example (8) considers the innovative development of DeepSeek as a tool through which China advances its geopolitical goals.

Example (8): DeepSeek’s development could also advance China’s geopolitical goals.

(NYT, February 2, 2025)

“goals” in Example (8) also represents the destination scenario of TRAVEL metaphor in NYT, aiming to constrain AI development in China. It can be seen that NYT endeavors to politicize technological issues when China is involved. Certainly, non-political goals are also described in NYT as well, which primarily focus on technological goals, commercial goals and health-related goals. These goals specifically include developing machines that can match the capabilities of the human brain and reducing the costs of AI training, etc. Some examples are shown as follows:

Example (9): But he said he believed that entirely new ideas were needed to reach the goal that Google and many others were chasing: a machine that can match the power of the human brain.

(NYT, December 19, 2024)

Example (10): “Everyone has been pursuing this goal because the costs of training AI was too high”.

(NYT, January 27, 2025)

(2)BUILDINGSANDCONSTRUCTIONMetaphor

BUILDINGS AND CONSTRUCTION metaphor in NYT is manifested by words such as “build*”, “blueprint”, and “underpin*”. Most of its relevant concordances emphasize the endeavors and investments made by technological companies such as Microsoft and OpenAI. Meanwhile, a very small number of concordances also highlight some regulatory actions concerning AI. For instance, in Example (11), the operational space of AI is construed as a constructed space that requires regulations through deliberate impositions of constraints.

Example (11): “Silicon Valley is investing billions to build these liberal guardrails to neuter the AI into forcing their worldview in the face of users and present it as ‘reality’ or ‘fact,’” Andrew Torba, the founder of Gab, said in a written response to questions.

(NYT, March 22, 2023)

(3)WARMetaphor

WAR metaphor in NYT can be found in two major scenarios including (1) the pre-war scenario illustrated by “risk*” and “threat*”, etc. and (2) the attack scenario evidenced by “fight*” and “strike*”, etc. The pre-war scenario focuses on the dangers of AI and government policies. Reports in NYT emphasize the dangers of AI in terms of its generation of online disinformation, as well as its threats to humanity, economic profits, and national security. Moreover, dangers of government policies mainly come from the U.S. government’s increased tariffs and its restrictions on AI chip exports, which threaten economic development. For instance, as to Example (12), AI DEVELOPMENT IS A WAR, in which the international expansion of Nvidia is construed as a target under the attack from the Biden administration’s restrictions on AI chips exports to China.

Example (12): The restrictions would threaten an international expansion plan that Nvidia’s chief executive, Jensen Huang, calls “sovereign AI”.

(NYT, December 29, 2024)

The attack scenario of WAR metaphor in NYT is typically realized by confrontational meanings such as “fight against” and “fight over”. Concordances conveying “fight against” position AI as a weapon of resistance against such risks as climate change, global warming and disease, etc. By contrast, concordances conveying “fight over” more emphasize the struggles over AI development, policy-making, and authorities among multiple stakeholders, as evidenced by the disagreements over dual visions for AI and controversies surrounding AI legislation. Particular attention in NYT has been paid to the metaphor that “AI LEGISLATION IS A WAR”. Since the Biden administration introduced new AI regulations, tensions have been intensified between the government and technological companies, as well as among administration officials affiliated with the Democratic and Republican parties.

5.3.3. Other Conceptual Metaphors with Significant Differences Distributed in News Reports on AI across ChinaDaily and TheNewYorkTimes

As shown in Table 3, as to the representative conceptual metaphors interpreted in the preceding two sub-sections, BUILDINGS AND CONSTRUCTION and FORCES metaphors occur significantly more frequently in CD than in NYT; whereas TRAVEL and WAR metaphors are significantly more prevalent in NYT. A closer examination of Table 3 also reveals four other metaphor types beyond those discussed above. This subsection therefore examines these less prominent but significantly differentiated metaphor types to further elucidate different cognitive representations of AI across CD and NYT.

(1)OtherConceptualMetaphorsSignificantlyOverusedbyChinaDaily

Compared to NYT, CD significantly employs PLANTS metaphor (x² = 7.80, p = 0.005 < 0.01) and PERFORMANCE metaphor (x² = 5.25, p = 0.02 < 0.05). PLANTS metaphor in CD can be evidenced by words such as “grow*”, “cultivate*”, “nurture*”, and “soil”, etc. These lexical items evoking the imagery of plant growth conceptualize AI and its applications as organic entities featured by expansion and sustainable development, while positioning China as a fertile soil for AI innovation. Moreover, the sustained growth of AI-related demands and social acceptance has been emphasized as well. To be more specific, users’ demands for AI models, workers’ acceptance of AI, and public attention to AI are all construed as the phenomena that “grow” progressively, which together suggest the self-expansive, cumulative, and long-term growth potential of AI development (Example 13).

Example (13): This technology is expected to usher in speedy growth in China and inject fresh momentum into a new industrial revolution, as major Chinese tech heavyweights step up efforts to develop their own AI-powered LLMs, experts said.

(CD, August 10, 2023)

PERFORMANCE metaphor in CD is primarily realized by the construction of “play* a(n)... role”, which conceptualizes AI development and its applications as forms of performance involving different participants, functional roles, and social implications. Relevant concordances mainly represent AI as playing a significant role across a series of fields such as healthcare (Example 14) and education (Example 15).

Example (14): Yan said that AI can also playapositiverole in enhancing hospital management and alleviating the lack of doctors.

(CD, May 19, 2023)

Example (15): AI-powered tools playanimportantrole in bridging urban-rural education gaps by expanding teaching resources and improving accessibility, Huang adds.

(CD, April 3, 2025)

(2)OtherConceptualMetaphorsSignificantlyOverusedbyTheNewYorkTimes

Compared to CD, NYT significantly employs ANIMALS metaphor (x² = 4.83, p = 0.03 < 0.05) and GAMES AND SPORT metaphor (x² = 8.77, p = 0.003 < 0.01). AMIMALS metaphor in NYT is represented by words such as “breed”, “dinosaur”, and “bull*”. In Example (16), the new generation of AI is represented as a new “breed” originally referring to a category of animals. This metaphorical expression naturalizes AI development as a process akin to biological evolution and reinforces a positive representation of AI as increasingly advanced and mature.

Example (16): Investors, pundits and journalists have talked up AI for years, but the new wave—the result of more than a decade of research—represents a more powerful and more mature breed of AI.

(NYT, January 7, 2023)

GAMES AND SPORT metaphor in NYT can be perceived by “race*”, “win*”, “game*”, and “compete*”, etc. Relevant concordances mainly emphasize those rivalries among AI products and technological companies (Example 17), as well as international competitions (Example 18) in AI development.

Example (17): Seeing the raw power of this phenomenon, companies like OpenAI, Google and Meta raced to get their hands on as much internet data as possible, cutting corners, ignoring corporate policies and even debating whether they should skirt the law, according to an examination this year by The Times.

(NYT, December 19, 2024)

Example (18): The moves dovetail with the Trump administration’s efforts to bolster the United States in an AI race against China, in part by cutting red tape.

(NYT, November 26, 2025)

The dialectical relationship between discourse and society suggests that the differences of conceptual metaphors employed in CD and NYT indicated above are largely influenced by social factors including media systems, national interests, and dominant ideologies.

Firstly, as to media systems, CD and NYT serve as the representatives of two different media systems in China and the U.S. CD is a Chinese official English-language outlet operating within a media system closely aligned with government policies. News reports in CD prioritize the construction of a national image by presenting the technological development and achievements in China. Therefore, relevant conceptual metaphors in CD tend to frame AI as a positive entity driving social progress and highlight the collective efforts made to enhance AI development across various sectors particularly at the national level. NYT situates itself within a liberal media system featured by market-oriented journalistic practices, in which news production involves balancing audience expectations with commercial considerations. Meanwhile, the liberal parties’ concerns about fairness, privacy, and humanistic values contribute to more emphases on the social risks and ethical dilemmas brought by AI in its corresponding metaphorical conceptualizations.

Secondly, as to national interests, CD serves to drive the development of new quality productive forces in China and construct China’s image as a responsible technological power by representing the implementation of national strategies. Such national strategies permeate the processes of industrial enhancement and international cooperation. In Chinese context, the “AI Plus” initiative proposed by China propels the integration of AI into various sectors to enhance productivity. Accordingly, PLANTS, FORCES, and PERFORMANCE metaphors are employed in CD to construct AI as a technological force characterized by growth potentials, empowering capabilities, and social implications. In global context, China has actively engaged in global AI governance to promote digital equity for developing countries and seek a bigger say on the global stage, which can be found in the BUILDINGS AND CONSTRUCTION metaphor that represents AI governance as a process requiring multi-stakeholder collaborations. NYT serves to sustain the technological leadership of the U.S. and stimulate the vitality of business markets by situating AI within the narratives of technological innovations, corporate competitions, and global technological rivalries. On the one hand, the U.S. maintains its competitive advantage in the global AI landscape through leading technological companies; thus, BUILDINGS AND CONSTRUCTION and TRAVEL metaphors are employed in NYT to portray AI development as a process driven by corporate investment and technological breakthroughs. On the other hand, AI is regarded as a catalyst to promote economic development and intensify corporate competitions. By employing ANIMALS and GAMES AND SPORT metaphors, NYT highlights the competitive relationships among market actors as they contend for technological resources and commercial interests. Furthermore, the WAR metaphor significantly overused in NYT places AI within technological competitions and geopolitical rivalries, indicating concerns of the U.S. with maintaining strategic advantages in the global AI race.

Thirdly, as to dominant ideologies, there exist a couple of distinctions between socialist collectivism and liberal individualism that influence the interpretations of conceptual metaphors across CD and NYT. More specifically, conceptual metaphors in CD are informed by socialist collectivism and a nationally coordinated development paradigm, thus emphasizing governmental guidance, multi-stakeholder collaborations, and the social implications of AI in promoting the shared well-being of humanity. By contrast, NYT is influenced by neoliberal individualism. Therefore, its uses of conceptual metaphors place more emphasis on market-oriented economics, as well as corporate innovations and competitions within the limited intervention of the U.S. government.