In the digital age, seeking answers online has evolved dramaticallyFor many years, the conventional practice of internet users was to turn to search engines, meticulously crafting queries laden with keywordsHowever, the rapid advancement of artificial intelligence (AI) technologies is transforming the search engine landscape dramatically, introducing a new paradigm often referred to as AI searchThis innovation liberates users from the necessity of painstakingly formulating keywords as they can now present their questions directly to AI systems, which not only provide answers but also synthesize and summarize the relevant information.

Wang Yunfeng, CTO of Worth Buying Technology, noted in an interview that "AI search is steadily encroaching upon the market share of traditional search engines, becoming a new mainstream entry point for traffic." This assertion resonates throughout the tech sector, where experts agree that generative AI is reshaping the search experience fundamentally

As we look forward to 2024, established players like Baidu and 360 are expected not only to embed AI capabilities within their existing search platforms but also to unveil autonomous AI search productsConcurrently, a wave of new competitors is emerging, joined together in a collective journey to explore the ultimate form of AI-powered search.

The transformation in user behavior is notableTraditionally, when engaging with standard search engines, users were required to refine their questions and distill them into clusters of keywords—a process that frequently resulted in a convoluted and less-than-ideal experienceComplex queries that necessitate multiple keywords could suffer from issues like misaligned priorities or automatic omission of terms, leading to inaccurate search resultsConsequently, users often found themselves sifting through countless pages of content, searching for clarity amid noise.

As articulated by 360's team, the technology underpinning traditional search engines is rooted in older natural language processing (NLP) algorithms, which struggle with longer search queries

• Why H-Shares of State Banks Face Scrutiny
• The Stock Market Celebration is Coming to an End
• Rate Cuts on Hold Amid Strong Economic Data
• Lincoln China: 2024 Initiatives and 2025 Outlook
• Carbon Neutrality: An ESG Investment Case

They indicated that the effectiveness of search plummets as query length increases, failing to accommodate more than 40 charactersIn stark contrast, AI search models, fortified by extensive data, are capable of accurately discerning user intent, adeptly processing search phrases that could be 4,000 or even 4 million characters in length.

According to Zhang Guangzhi, a director at the Beijing Big Data Association and chief scientist at Rongxin Data, “In the past year, the most significant shift in search has been the capacity of large model search products to deliver precise and highly valuable results, directly presenting the content users desire, rather than requiring them to laboriously navigate through results.” This evolution has contributed to a more streamlined and user-friendly relationship between individuals and the information they seek.

The fundamental value of a search engine—a connective tissue between users and the information they pursue—has been radically redefined

Zhou Xiaopeng, vice president of Alibaba's Smart Information Division, emphasized in an interview that AI search is revolutionizing the way information is consumed, drawing connections closer to the ideal of immediacy and accessibilityThis capability, he argued, provides a robust foundation for the broad potential of AI search products in the future.

Given its simplicity and alignment with natural human inquiry, AI search has garnered increasing popularity among usersFor instance, Han Hui, a freshman at a university in Hubei Province, shared that he frequently employs AI search products like Doubao and Wenxiaoyan to assist with various coursework assignments.

“Everyone in our dormitory uses AI search, and we often recommend effective AI tools to one another,” Han remarkedThis observation underscores a significant trend; third-party research firm Gartner predicts that by 2026, user traffic directed to conventional search engines may dwindle by as much as 25%.

As the landscape evolves, the entry of new players into the market and the emergence of independent AI search products are palpable

Traditional search engine companies are actively integrating AI capabilities, as demonstrated by the way Baidu and 360 now offer “AI smart responses” as their top answers when users pose queries.

In 2024, a discernible shift will be the proliferation of autonomous AI search technologiesDespite the incorporation of AI functionalities, companies like Baidu and 360 have opted to create stand-alone AI search products, amplifying their competitive edgeBeyond these established entities, numerous newcomers such as Alibaba, ByteDance, Bilibili, and Kunlun Wanwei are launching AI search products that meld seamlessly with their existing products and ecosystemsA multitude of startups, as well as specialized companies, are also stepping into the arena, cultivating a rich variety of AI search offerings tailored to niche markets.

The rise of AI search signifies not merely an enhancement of search algorithms but a transformative overhaul of how users retrieve information

The entrance of new competitors is gradually dismantling the so-called “oligopolistic” nature of the traditional search market.

Across various AI search platforms, distinct trends are emerging based on their specific strengthsFor instance, ChatGPT is known for its ability to generate summarized answers, while Perplexity focuses on curating search results through AI, and Genspark emphasizes producing rich content including comprehensive reports based on guided keywordsZhang observed, “Different AI search solutions cater to unique customer demographics, and users might gravitate towards platforms that provide precise answers or analytical reports, depending on their specific inquiries.”

Additionally, platforms with a more private community focus, such as Xiaohongshu and Zhihu, are introducing specialized AI search productsThese platforms boast a wealth of proprietary specialty content and user-generated material, which not only serves as valuable datasets for AI search but also accelerates the route to high-quality content access, thus enhancing search efficiency

Various reports suggest that the introduction of these niche AI search products may gradually siphon users from traditional search engines, potentially leading to the rise of breakthrough applications.

In this evolving landscape, technical advancements serve as a catalyst for shifting the conception of search itselfAI search has rapidly evolved over the past couple of years, with 360's representatives explaining that the development can be distinctly categorized into three progressive phasesThe 1.0 phase, spanning late 2022 to early 2023, represented a fusion of traditional keyword searches with LLM (large language model) summaries, maintaining a fundamental reliance on existing paradigmsThe 2.0 phase, emerging from mid-2023 into late 2024, begins to reflexively integrate large model thinking into core search mechanics, introducing dialogic exchanges and collaborative mechanisms among various models, enhancing the system's intelligence and noise-filtering capabilities

The projected 3.0 phase commencing in late 2024 anticipates the integration of multi-modal content creation engines featuring profound interoperability across different contexts.

It is essential to recognize that various phases coexist in the current landscapeNevertheless, industry insiders are confident that the progressive evolution of AI search mandates an inevitable advancement toward the 3.0 phaseReflecting upon the past year, it is clear that the conception of the search box in the AI era has transcended the limitations of traditional searches; no longer constrained to mere text queries, AI search now promises a multifaceted experience that includes text, image, audio, and video contentThis evolution positions AI search as an all-encompassing intelligent assistant capable of handling a full spectrum of functions, from searching to storage, integration, extraction, and content creation

Indeed, some AI search applications are already experimenting with features like “ask by taking a photo.”

Zhang observed, “AI search has progressed from merely providing Q&A formats to now referencing source materials and ultimately offering multi-dimensional organization of answers, such as outlines and expansive reports.” Predictions suggest that by 2025, general-purpose AI searches will yield even more accurate results, evolving into a platform that excels not only in search but incorporates peripheral functionalities designed to meet comprehensive user needs.

Moreover, emerging from a perspective at 360, it was reaffirmed that the journey of search extends beyond user queriesFrequently, users seek a "production loop" that transitions seamlessly from problem-solving to content creation, culminating in an integrative workflowBy the end of 2024, a range of AI search products is expected to integrate or consider integration with neighboring service capabilities, further enriching the ecosystem.

Chairman and CEO of Kunlun Wanwei, Fang Han, encapsulated the current state of AI, acknowledging that while the progression of training-related scaling laws appears to be slowing, a new approach—based on slow and thoughtful reasoning—has begun to take form