ARTIFICIAL intelligence (AI) is transforming how advertisements are created and delivered in China, reshaping the marketing sector that has long driven profits for Internet companies.
Leading tech firms and top online advertising platforms such as Tencent, Alibaba Group, Baidu, ByteDance and Kuaishou Technology are speeding up efforts to boost advertising productivity through advanced AI models.
Moving beyond targeted placements, these platforms now claim that they can generate large volumes of ad content with a single click, with algorithms optimising both selection and distribution.
“This is the era of intelligent advertising, with models powering everything from content creation to ad placement,” a marketing product manager at a short video platform told Caixin.
With AI-generated content (AIGC) at the forefront, ad automation has reached new heights, as systems identify top-performing content from a vast array and pair it with ideal bidding and placement strategies, the person added.
AI is helping advertisers achieve their ultimate goal: reaching targeted users with personalised ad content.
BT in your inboxStart and end each day with the latest news stories and analyses delivered straight to your inbox.
Shen Chengang, founder and chief executive officer of digital marketing solutions provider Meetsocial Group, said that AI is transforming digital marketing by predicting and simulating ad performance; guiding businesses in choosing channels and methods for ad placement; enabling real-time strategy adjustments based on AI feedback; and tailoring ad content to specific users and contexts.
“In the near future, AI will be able to automatically generate personalised ads in real time based on context predictions, user interests and product characteristics,” Shen added.
China’s Internet user growth has slowed, signalling the end of a business boom driven by its huge population. According to Big Data intelligence services provider QuestMobile, from June 2023 to June 2024, the number of monthly active Internet users in China rose by only 22 million, with growth slowing from 2 per cent to 1.8 per cent.
Faced with tightened budgets, advertisers are focused on maximising their return on investment, seeking more cost-effective ways to drive more traffic with less spending.
The adoption of AIGC has sharply reduced ad production costs, bringing the price of a single performance ad down from several hundred yuan to under 1 yuan.
Short video platform Kuaishou reported that with a one-stop video editing tool, the cost of creating a single video ad has fallen from more than 200 yuan (S$37.2) to just 0.47 yuan. ByteDance claimed that similar tools have increased ad content production capacity four-fold.
These reduced costs are fuelling ad spending, allowing China’s Internet advertising market to grow rapidly despite weakened consumer demand.
In the first half of 2024, the market expanded by 11.8 per cent year on year to reach 351.4 billion yuan, according to QuestMobile. For the whole of 2023, it hit 714.6 billion yuan, a 7.6 per cent increase from 2022 and a 52 per cent rise from pre-pandemic levels in 2019.
AI-driven performance gains are beginning to yield results. In H1, companies such as Tencent and Kuaishou, which have introduced AI advertising tools, reported that AIGC adoption has spurred advertisers’ spending and boosted revenues.
Tencent saw ad revenue growth of around 20 per cent in each of the first two quarters, while Kuaishou’s ad revenue rose by 25 per cent in the first half of the year. AI models now support the entire ad creation and distribution process, producing more than 100,000 short video ad assets daily, Kuaishou said.
Now advertising professionals are beginning to feel anxious about their jobs. An ad optimisation specialist told Caixin that in the past, ad systems required manual setup for campaigns and targeted keyword selection, but AI has now streamlined these tasks. “In the future, it could become even more integrated and accessible, making it easy for anyone to use,” he said.
AI-powered growthFor China’s top 10 Internet companies by market value, nearly 30 per cent of total revenue stems from advertising. With overall growth slowing, AI-driven performance boosts are essential.
In 2023, ByteDance, which owns short-video sensation TikTok and its Chinese version Douyin, saw its total revenue surpass Tencent. ByteDance has deeply engaged in AI for Douyin and its news aggregator Toutiao, which builds user profiles by analysing content-viewing habits and matches ads through personalised recommendations.
Ads are embedded within short videos or articles, increasing exposure as users spend more time on these platforms.
ByteDance’s ad business is one of the most dynamic. In early 2023, its ad platform Ocean Engine introduced a “large-scale, real-time prediction system”, which forecasts user interests based on past behaviour. Leveraging multimodal and large language models, it helps advertisers automatically generate ad scripts.
Rival Kuaishou this year unveiled its text-to-image generation model Kolors, which has been applied across Kuaishou’s short video, advertising and live-streaming e-commerce segments.
At the 2024 World Artificial Intelligence Conference, Kuaishou said that nearly 20,000 merchants had adopted its large-model capabilities for “intelligent operations” in H1, with daily ad revenue from AIGC-driven content passing 20 million yuan in June. With a daily ad revenue average of 187 million yuan, nearly 11 per cent of Kuaishou’s ad income is now AIGC-driven.
Tencent has also made advertising the focal point of its large-model applications. Powered by its Hunyuan large language model, Tencent launched its Ad Delivery System 3.0, which tracks user behaviour over longer periods to better understand consumer intent, attracting increased ad spend.
According to an ad optimisation specialist, the approach is about “content finding its audience”, as the system combines industry data with user behaviour to tailor ad recommendations.
For advertisers, the goal is to capture precise targeted traffic. In August 2023, Alibaba’s marketing platform Alimama introduced a new ad system that uses generative AI to simplify content creation while using proprietary large models to intelligently match ads with potential consumers.
Uncertain profitDespite the strong performance, AI-driven advertising is still struggling to achieve profitability or attract significant client numbers. High research costs and unpredictable outcomes have kept returns on AI ad investments in the red, according to a marketing manager from a short video platform.
Zhou Jian, GroupM China’s brand safety director, said that the top 10 Internet platforms account for 95.9 per cent of China’s digital ad spending, shaping trends in media investment. While AI is deeply integrated into advertising, its transformative potential remains uncertain.
Meanwhile, AI’s reliance on industry-best practices has led to content homogenisation, which could degrade the quality of the ad ecosystem.
A former ByteDance product employee said that AI often learns from popular ad formats, which – while effective in the short term – may shorten content lifespan and reduce creative diversity.
The growing reliance on AI-generated content for advertising has also led to concerns about copyright and user privacy.
Regulatory bodies in Europe and other regions are increasing their scrutiny of how tech companies manage AI training data. Meta Platforms and Google, both heavily reliant on ad revenue and actively developing AI applications, are facing tougher regulatory scrutiny.
Liu Ting, head of data and technology consulting at GroupM’s data unit Choreograph, said that the use of generative AI in advertising poses risks such as data breaches, copyright infringement, data contamination and the exposure of sensitive information.
Job displacementThe rise of AIGC in marketing and advertising has affected traditional industry practitioners. According to data from the State Administration for Market Regulation, the number of advertising companies registered in China dropped to 17,000 in 2023, a decrease of nearly 1,000 from the previous year.
According to GroupM’s Zhou, AI’s rapid production capabilities and big data advantages have disrupted supply of and demand for talent. The quick generation of content, bulk copy production and the use of digital avatars to replace live hosts have reduced time and labour costs.
However, he added that AI’s impact on jobs may not be as pronounced in the short term as some fear.
While demand for certain roles has fallen, expectations for advertising professionals are rising. An ad optimisation specialist said that as content creation tools become more user-friendly, the potential for roles in editing and copywriting to be replaced increases, leading to a greater emphasis on data analysis and the opportunities that entails.
Research from iResearch shows that in 2023, 48 per cent of advertisers had begun using AIGC technology in online marketing. A report from CTR Market Research found that the penetration rate of AIGC in marketing rose from 69.6 per cent in 2023 to 78.9 per cent in 2024, with nearly 80 per cent of advertisers planning to adopt AIGC.
Despite this, 73 per cent of advertisers believe that AIGC cannot replace humans in making critical decisions. CAIXIN GLOBAL
{Categories} _Category: Applications{/Categories}
{URL}https://www.businesstimes.com.sg/international/global/china-ai-taking-internet-advertising-storm-what-cost{/URL}
{Author}unknown{/Author}
{Image}https://cassette.sphdigital.com.sg/image/businesstimes/2a20168ddb6228b5b24c186f73700d8fd630da23ed389e1e98db4e39febb50d7{/Image}
{Keywords}Global{/Keywords}
{Source}Applications{/Source}
{Thumb}{/Thumb}
