A Comprehensive Understanding of the AI War in the United States – The Battle Between the "Big Five Tech Giants" and the "Three Little Dragons of AI"

The American technology industry is undergoing an unprecedented arms race in artificial intelligence, with the competitive landscape between traditional tech giants and emerging AI companies being reshaped.

The five major tech giants—Apple, Google, Meta, Microsoft, and Amazon—are facing challenges from the "AI three little dragons" such as OpenAI, Anthropic, and xAI, as each company vies for dominance in this new era that could redefine the entire tech industry.

Meta CEO Mark Zuckerberg's recent frenzied hiring spree reveals the intensity of this competition. According to a previous article by Wall Street Insight, CEO Zuckerberg is engaged in a mad talent grab, personally reaching out to hundreds of researchers to join his newly established "superintelligence" lab, with offers for individual talents reaching as high as $100 million, highlighting the company's anxiety in the AI race.

The game between traditional tech giants and new AI players is not only about technological leadership but also directly impacts their core business models. For Google, conversational AI like ChatGPT poses a disruptive threat to its search business; for Apple, AI serves more as a supplementary tool to enhance the experience of its devices. This differentiated impact is reshaping the competitive landscape of the entire tech industry.

Apple: Hardware Advantages Remain, Need for Deeper AI Collaboration

Apple's performance in the AI field has been disappointing, with the Apple Intelligence project encountering significant setbacks. Although the company possesses basic capabilities in large language models on devices and private cloud computing infrastructure, it has not yet reached the cutting edge in terms of models and products.

However, Apple's core business is not directly threatened by AI. From a consumer perspective, AI applications like OpenAI and Claude still need to be used on the iPhone or accessed through a browser, and development tools like Cursor also need to run on a Mac. Apple's local large language models can provide differentiated features for applications on the Apple platform, and the company's unique access to consumer data allows it to build practical and scalable personal semantic indexes.

Apple's current situation is similar to Microsoft's during the internet era: while people used the internet on Windows PCs, it was the internet that created the mobile era, surpassing the PC paradigm. By accepting its failures in the mobile space, Microsoft built complementary businesses (cloud computing) and regained a favorable position in the AI era. Apple should also emulate this strategy and deepen its collaboration with OpenAI.

Apple should focus on building the best hardware for the AI era, including AI-driven devices like the Apple Watch, HomePod, and glasses, and in the long term, should invest heavily in robotics and home automation. In terms of quality and scalable production of consumer hardware, Apple remains unmatched

Google: Leading in Infrastructure but Facing Threats in Search Business

Google's performance in the AI field has exceeded expectations, with its infrastructure being world-leading in many aspects. The company has achieved complete integration from chips to networks to models, with the Gemini model performing exceptionally well in evaluations, particularly leading competitors in media creation fields such as video generation.

Google's most significant advantage may be its data. The Veo video generation tool can leverage YouTube video resources, the scale of which is unimaginable. Google's large language models benefit not only from the company's leading position in web indexing but also from other forms of data collected over the years, such as scanned books and archived research papers.

Google also has distribution channels, particularly the Android system. The potential for providing integrated devices and cloud AI experiences is substantial, making Android's challenge to Apple's dominance in the high-end market the best opportunity. The cloud computing business is even more promising, where Google's infrastructure and model advantages (especially in pricing) can truly come into play without worrying about protecting revenue or reactivating aging product capabilities.

However, the disruptive impact of AI on Google's core search business remains a major concern, which is the biggest challenge Google faces. Google is improving the search experience through AI search overviews and developing search funnels to systematically evolve search for AI, transforming AI from a disruptive technology into an enhancement tool.

Meta: Strategic Position Solid but Execution Faces Challenges

Meta's positioning lies between Apple and Google, and its core strategic positioning seems more solid: personalized content and generative advertising should enhance Meta's core social media business. At the same time, AI is likely to be key to Meta's return on investment in VR/AR.

However, risks exist. The scarce resource that Meta is competing for is attention, and large language models are already consuming vast amounts of attention and continuously expanding. Some argue that this makes chatbots a threat to Meta as significant as they are to Google.

Note: In the AI field, attention mechanisms are key technologies in model training and inference processes, especially when dealing with large-scale language models, where the performance and efficiency of attention mechanisms directly affect the training speed and effectiveness of the models.

From Zuckerberg's frantic hiring, these risks may be greater than expected. While Zuckerberg has proven willing to invest billions in an uncertain future, the speed and urgency of this AI hiring strongly indicate that the threat to the company's core business exceeds expectations.

Worryingly, Meta seems to lack a clear direction in AI. Zuckerberg's thoughts on how generative AI will impact the company's business appear to be underdeveloped. This large-scale hiring reflects the leadership's awareness of the seriousness of the issue.

Microsoft: Position Remains Strong but Faces New Challenges

Microsoft's position appeared unshakeable in January 2023 and still holds a favorable position, but the situation is more nuanced than it was then: The relationship between Microsoft and OpenAI is becoming increasingly tense, recently evolving into a threat from OpenAI to file an antitrust complaint if Microsoft does not relinquish future profit rights and agree to the profit restructuring proposed by OpenAI. The current agreement will expire in 2030.

Nevertheless, Microsoft still holds significant advantages: Azure is the exclusive non-OpenAI provider of the OpenAI API, which not only retains enterprise customers but is also a powerful attraction in itself. Microsoft should prioritize ensuring Azure's advantageous position in negotiations with OpenAI, even if it means giving up certain rights to OpenAI's business.

Microsoft should also deepen its relationships with other model providers. The pre-recorded video appearance of xAI CEO Elon Musk at the Microsoft Build conference is a positive signal, and Microsoft should consider investing to help ensure xAI continues to pursue cutting-edge developments in AI models. The company has also made a small investment in Mistral and should consider helping to fund the Llama project.

Amazon: The Potential of AWS as a Latecomer

Over the past two years, Amazon's position has become more optimistic:

First, AI is not disruptive to any of Amazon's businesses; on the contrary, it will benefit from it. The increase in AWS usage is evident, and Amazon.com may also be a huge beneficiary of customers using AI for product recommendations.

Second, the partnership between AWS and Anthropic seems more stable than the collaboration between Microsoft and OpenAI. Anthropic's lack of a strong consumer business makes maintaining a vendor-type relationship with AWS more feasible.

Third, AWS's early investment in Bedrock was an early bet on AI optionality. The company's investment in Trainium provides similar benefits for the future of AI chips.

The "Three Little Dragons" of AI Each Show Their Skills to Seek Breakthroughs

The three major AI model manufacturers are adopting different strategies to establish and maintain their market positions, and their partnerships with large tech companies are also characterized by distinct features.

OpenAI has established a dominant position in the consumer AI space, with the success of ChatGPT making it an "unexpected consumer technology company." This positioning creates a fundamental conflict with any entity seeking to have customer relationships, including Microsoft and Apple. However, these two companies may have no choice but to collaborate with OpenAI:

Microsoft has already invested too deeply, while Apple may find that OpenAI quickly becomes the most compelling reason to purchase an iPhone. The key question facing OpenAI is when and whether it will launch an advertising model to supplement its subscription business.

Anthropic has missed the consumer market but has built a strong position within the developer community, with its programming-focused strategy generating a robust API revenue stream. The company's training partnership with Amazon and its Trainium chips may not be optimized in the short term, but in the long run, it could mean significant cost savings. More importantly, Amazon is now a deep partner of Anthropic, and this relationship is more stable than the collaboration between Microsoft and OpenAI xAI is in the most difficult situation. The company's insistence on having its own infrastructure feels more like a burden than an asset. xAI is a company that everyone hopes exists, serving as an alternative to balance OpenAI and Anthropic. The company should particularly seek investment from Microsoft.

From a certain perspective, it makes sense for xAI and Oracle to be partners: xAI can utilize infrastructure partners, while Oracle can leverage differentiated AI products. The problem is that they may exacerbate each other's challenges in acquiring customers