The conversion rate of AI-generated advertising materials is 20% to 30% higher than that of manually created materials, according to recent findings discussed by industry leaders at a 36Kr event. This shift, coupled with the decreasing costs of AI-driven content creation – now reaching just a few cents per piece – is redefining efficiency in the advertising and marketing sectors.
Jiang Peng, Vice President of Kuaishou Technology and Head of Commercial Algorithms, and Kazik, CEO of Virtual Reality Media and operator of “Digital Life Kazik,” explored the implications of this transformation. The discussion centered on the rapid evolution of AI, the rising popularity of tools like OpenClaw, and the impact on content creation and advertising ROI.
Both executives identified DeepSeek R1 as a pivotal moment in the AI landscape. Kazik stated that DeepSeek R1 demonstrated that Chinese models could achieve performance levels comparable to OpenAI’s offerings, while significantly reducing costs. This breakthrough sparked widespread discussion across the country. He as well highlighted Nano Banana as a revolutionary tool for designers, streamlining workflows and reducing the value of previously essential components within image editing software.
Jiang Peng echoed the sentiment, pointing to the profound transformation of the content industry through AI. He noted that the ability of AI to create content has advanced rapidly, fundamentally impacting the entire content ecosystem. He also emphasized the growing importance of AI agents, predicting that 2026 will see agents become a primary focus of the AI industry, shifting the focus from simply answering questions to completing tasks.
However, Jiang Peng also cautioned about the costs associated with AI, citing an example of a colleague in Silicon Valley who found the expense of using an AI agent as a digital assistant exceeded their own salary. This highlights the increasing require for the industry to address both the costs and monetization of AI.
The discussion turned to the surge in popularity of OpenClaw. Kazik explained that OpenClaw, like DeepSeek R1, gained traction through a process of cognitive leaps in public perception. He argued that ChatGPT made people aware that “AI can converse,” DeepSeek R1 demonstrated that “AI can think,” and OpenClaw introduced the concept of the “agent” to a broader audience. He contrasted this with earlier agent products like Manus and Claude Code, which were cloud-based services that felt like rentals, while OpenClaw allows users to “hold” the agent within their own systems, fostering a sense of ownership.
Jiang Peng agreed, stating that OpenClaw allows ordinary users to experience the practical benefits of AI, furthering public understanding of the technology. He noted that OpenClaw largely aligns with agent technologies developed over the past two years, with its unique value lying in its accessibility to everyday users.
Regarding the effectiveness of AI-generated advertising, Jiang Peng revealed that on the Kuaishou platform, the cost of creating a single ad has fallen to just a few cents, while conversion rates have increased by 20% to 30% compared to manually created materials. He emphasized the importance of personalized strategies, tailoring materials to specific product features and user characteristics.
Kazik highlighted the cost savings in video production, stating that AI-generated TV commercials can be produced for a fraction of the cost of traditional methods – potentially 10,000 to 20,000 yuan per minute compared to 100,000 to 300,000 yuan for traditional production. He also noted the significant reduction in both personnel and time costs.
Jiang Peng explained that Kuaishou’s “generative reinforcement learning for bidding” technology has increased advertising revenue by 3% to 4%, improved advertising capacity for some clients by 20% to 30%, and improved ROI. The platform’s automation level for advertising placement has reached approximately 90%, with clients only needing to specify their ROI goals and the products they wish to promote.
Kazik differentiated between quantitative advertising and quantitative transactions, arguing that advertising platforms have a global perspective and aim to maximize benefits for all clients, while quantitative transactions are more competitive.
Looking ahead, Jiang Peng predicted continued competition surrounding AI tools, but emphasized that the focus will increasingly shift towards efficiency and return on investment. He stated that companies want users to genuinely utilize AI tools, not simply take advantage of promotional offers.
Kazik echoed this sentiment, noting that cognitive development in the public’s understanding of AI occurs in stages and cannot be forced through financial incentives alone. He emphasized that establishing recent cognitive positions – such as AI’s ability to converse, think, or act as an agent – is crucial for building lasting user engagement.
