Source: Caijing News Author: Shi Zhengcheng

After the closing bell on Tuesday, Apple's stock price surged more than 7%, with a market cap of $3.18 trillion, breaking the historic high set in mid-December last year and hitting the first historic high of the year for consumer electronics giants.

On the news front, what still stimulated the rise in stock prices was yesterday's WWDC conference. The company unveiled the AI strategy that investors have been waiting for, as well as its cooperation with OpenAI. Although many big names in the tech industry have expressed that Apple Intelligence exhibits Apple's software development skills and philosophy, still a few unexpected breakthroughs are lacking.

However, at least some Wall Street investors seemed satisfied with yesterday's press conference based on this big bullish candlestick on Tuesday morning.

Focusing on the iPhone upgrade cycle.

One thing that most analysts who are bullish on Apple these two days have in common is that they see Apple Intelligence as a driver of the iPhone's upgrade cycle.

According to a previous report by Cailian Press, Apple's AI strategy will focus on "localization"-mainly letting the chips of mobile phones and computers bear the computing function. This ensures user experience (speed) while maximizing user privacy protection during the AI usage process.

Even Musk, who was arguing about "banning Apple from entering companies" after the conference, still mainly attacked Apple's ability to call external AI service providers in the system, and the only official partner currently announced is Musk's nemesis OpenAI.

Apple officially announced that only the iPhone 15 Pro and iPhone 15 Pro Max can use Apple Intelligence among all iPhone models already on the market. Any Apple chip models released in recent years are OK on the iPad and Mac side.

Wedbush analyst Dan Ives said that Apple launched multiple new software enhancement features at the moment of "showing its strength", which will enable the company to enter a strong product cycle for the next 6-9 months, including the launch of this year's autumn iPhone 16 models.

Wamsi Mohan, an analyst at Bank of America, also pointed out that the limitations of Apple Intelligence may trigger the upgrade cycle of AI function smartphones, as consumers hope to gain productivity by upgrading new functions. Toni Sacconaghi, an analyst at Bernstein, also believes that Apple will see a stronger upgrade cycle on the iPhone 16.

In addition, Gil Luria, an analyst at D.A. Davidson, also put forward another point of view: Apple has become the first company to provide "meaningful AI agent capability". After the upgrade, Siri can perform operations across applications and has "screen awareness" function.

Under the drive of "AI is coming" and the announcement of the largest share buyback plan in US history in early May ($110 billion), Apple rose rapidly by 13% in May this year, achieving its largest monthly increase since July 2022. However, Apple's cumulative increase this year is only 6%, which is far lower than the 14% increase in the Nasdaq.

Apple executives explain the AI vision.

After the broadcasted press conference ended yesterday, Apple's executives also attended various activities of WWDC, emphasizing the difference of "Apple AI".

Apple's AI strategy director Giannandrea and software development supervisor Federighi emphasized yesterday that Apple's AI strategy is different from that of competitors and those companies that build AI for a wide range of products. Apple focuses on hardware and personal data that AI may use. Federighi emphasized that Apple believes "the role of artificial intelligence is not to replace our users, but to empower them".

Apple's AI vision is not to create a super-powerful "big model", but a series of small models that can run locally. These functions are packaged as "Apple Intelligence". Apple's strategy is to ensure that the "small models" work only on the tasks they are good at, rather than providing users with an open chatbot interface.

Giannandrea emphasized that Apple believes the right approach is to have a range of models for different usage scenarios. He revealed that as part of Apple Intelligence, the company is committed to creating a model with a parameter size of 3 billion.

By comparison, in 2020, GPT-3 had 175 billion parameters. The more parameters a model has, the more computing power and memory it requires to run.

As for the cooperation with OpenAI, Apple also quickly expressed the possibility of introducing more models in the future. Federighi said that some customers may want to embed fsg medical serv a or legal AI models in Apple products, or maybe some models from Google. He emphasized: "Apple looks forward to future cooperation with Google Gemini, although there is no announcement yet, this is indeed our direction."

Editor/Lambor