Although Apple has avoided using the abbreviation "AI" when discussing its products for many years, today it is no longer possible to avoid it. However, as a follower in AI, Apple faces fierce competition from major technology companies.

How does Apple respond to its competitors' continuous moves?

Since the generative AI boom caused by OpenAI at the end of 2022, AI has been the biggest topic in the technology industry, driving the market cap of chip maker Nvidia to reach $3 trillion and causing significant shifts in global strategy for companies like Microsoft, Google, and Amazon, who are all vying to incorporate AI into their core services.

OpenAI, the leader in generative AI models, has seen a significant increase in revenue since the company released its new model, ChatGPT-4o, which includes voice conversation functionality. The company also announced that a new generation model is in development.

As direct competitors of Apple in the field of electronic devices, Google and Microsoft have already made big moves in the AI PC field.

Microsoft announced that Windows 11 AI PC will be gradually launched in the devices of its partners, such as Dell and Lenovo, starting from June 18, which is 58% faster than the Apple MacBook Air with M3 and provides real-time subtitles in over 40 languages with a computing frequency of over 40 TOPS/second using its new Recall feature.

Google has announced the addition of new AI features to its high-end Android laptop Chromebook Plus product line, entering into direct competition with Microsoft and Apple in AI PC. Google also has its own Gemini model, which at first was overshadowed by ChatGPT but is narrowing its gap with ChatGPT-4o with continuous iteration.

With the Apple Worldwide Developers Conference (WWDC) set to take place next week in 2024, investors and customers are hoping to see what AI features the iPhone maker will offer in response to intense AI challenges.

Apple CEO Tim Cook has hinted at "major plans", and everyone is watching to see what Apple will achieve in AI. Currently, Apple has over one billion iPhones in use, and Wall Street hopes to hear about which AI features will make iPhone more competitive against Android competitors and how the company will demonstrate its investment in developing proprietary chips.

Investors are more interested in companies that show a clear AI strategy and vision. Nvidia's stock has tripled in the past year, and Microsoft has actively incorporated OpenAI into its products, rising 28% in the past year. In contrast, Apple has only risen 9%, and its market cap has been surpassed by the aforementioned two companies.

Analysts say this is the most important event for Cook and Apple in over a decade. "AI strategy is the missing piece in Apple's growth puzzle, and this event needs to be eye-catching, not indifferent," said one analyst.

It is expected that Craig Federighi, Apple's software head, will talk about the practical application of Apple AI. Some analysts predict that Apple will reveal its long-term vision of implementing generative AI throughout its diverse ecosystem of personal devices. "The impact of generative AI on Apple's business is the most far-reaching in the entire technology sector, and Apple has a clear opportunity to bring generative AI functionality to its billions of consumer devices," said an analyst.

Cooperating with OpenAI or competing with it?

Last month, OpenAI launched the voice mode of its AI software ChatGPT-4o. When ChatGPT-4o was first launched, some pointed out that it should be the goal of Siri's update.

Some analysts believe that Apple may work with OpenAI to upgrade Siri. There are also reports that Apple is discussing getting chatbot technology from other companies, including Google and Cohere. But Apple has refused to comment on its cooperation with OpenAI.

However, some analysts believe that Apple's new Siri will not directly compete with fully functional chatbots, but will improve its existing functionality and pass any questions that only chatbots can answer to its partners. This is similar to the way Spotlight and Siri currently work. Apple's system will try to answer questions, and if it cannot, it will turn to Google. The agreement brings in $18 billion in revenue for Apple each year.

Moreover, Apple may also avoid embracing cooperation with OpenAI or chatbots. One reason is that chatbot malfunctions can lead to embarrassing headlines and may undermine the company's emphasis on user privacy and control of personal data. OpenAI's technology is based on web crawls, and ChatGPT user interactions are used to improve the model. Such technology may violate Apple's privacy principles.

In addition, large language models like OpenAI still have problems with inaccuracy or "hallucination," as well as unresolved copyright issues, all of which are concerns for Apple executives who hope to avoid.

Some senior figures in the tech industry believe that even if Apple and OpenAI collaborate, it is likely to be just a stopgap measure. Apple has long been incorporating external technology into its own and is committed to developing its own capabilities, such as using its own chips to replace those of Intel.

The source also revealed that Apple's AI strategy is far broader than just chatbots. It plans to use large-scale language models to drive secretly developed robotic devices, including a desktop robotic arm with a large display similar to the iPad and a mobile robot that follows users and handles household chores on their behalf. Additionally, Apple hopes to equip AirPods with cameras and AI functions.

Meanwhile, the possible collaboration between Apple and OpenAI has also raised concerns among Microsoft executives.

Analysts believe that Microsoft hopes to use OpenAI's technology to enter the consumer market, not just the enterprise market, and the transaction between OpenAI and Apple could harm these opportunities, pitting Apple's AI capabilities directly against Microsoft's.

However, on the bright side, Microsoft may make more money from the Apple and OpenAI collaboration, as it will receive more revenue from cloud computing.

Efficiency vs Scale.

Apple may also differentiate itself from competitors in the emphasis of its AI model route.

Cook previously emphasized Apple's proprietary chips' advantages in running AI models. "We believe in the transformative power and vision of AI and believe we have some unique advantages to make us stand out in this new era, including Apple's distinctive seamless hardware, software and service integration, groundbreaking Apple Silicon and our industry-leading neural engine, as well as our steadfast commitment to privacy."

Analysts predict that Apple's WWDC keynote will focus on the capabilities of its functions and devices, as well as its generative AI models running on those devices.

In April of this year, Apple released research on its AI model called "Efficient Language Model," which can run on smartphones. Microsoft is also researching the same concept. One of Apple's "OpenELM" models has only 1.1 billion parameters or weights, much smaller than OpenAI's 2020 GPT-3 model's 175 billion parameters and even smaller than one version of Meta's Llama's 70 billion parameters.

In the paper, Apple researchers benchmarked the models on MacBook Pro laptops running the Apple M2 Max chip, showing that these efficient models do not necessarily need to be connected to the cloud. This can increase response speed and provide a layer of privacy protection because sensitive questions can be answered on the device itself rather than returning to Apple's servers.

Media reports suggest that some of the features built into Apple software may include providing users with summaries of missed text messages, generating images of new emojis, writing code in the company's development software Xcode, or composing email replies. Apple may also decide to load the M2 Ultra chip into its data centers to handle more AI queries that need more computing power.

Editor/Emily