Apple is taken in its own dangerous wrestling-22

Update, Sunday March 16, 2025: This article has been updated with more details on the next Apple steps to approach assembly criticisms around Apple Intelligence.

The weak efforts of Apple AI were highlighted this week with an announcement that certain improvements in Apple’s intelligence in Siri are be delayed until 2026. This defect in delivery of the features presented in the keynote WWDC 2024 saw the Apple commentator noted John Gruber serves a dazzling disassembly Apple efforts to date.

He is not the only one.

While competition rupture in the distance fueled by AI, what challenges is confronted with Apple and what options are open to it?

Apple Intelligence Catch-22 challenge

First, recognize that artificial intelligence – in particular a generative AI – is not only a fashionable word used to sell the latest technologies but a technique that has immersed the landscape. Sometimes it is at the front and in the center, like the co -pilot of Microsoft and the Gemini of Google. AI is more subtle but clear in other areas, such as backfilling images and cultures in Photoshop. In more areas, it is almost invisible but accepted; The grammar tool is a good example.

If AI is present on several ecosystems in countless modes, not having a competent AI is a fundamental weakness. Apple is already left behind; Without finding a mastery of AI, he can never make up for it.

Then, you have the cornerstone of the majority of AI tools used today … Large data sets used to train the different models so that they can offer sufficiently robust outputs that meet expectations. Companies that have collected data on their user basis (for any reason) have ready access to these large data sets.

Apple has long been a vocal supporter of the sacred nature of the personal data of its users. He keeps as much as possible on the handset of a user, figures the data he attracts in the cloud so that he is out of reach of his own staff, and will fight to preserve these principles in courts against national governments if necessary. The modification of this principle would rock the Apple ecosystem to its heart.

Endless loop of the Apple Intelligence

Apple lives in a world where AI will be table issues; AI needs large training sets to which Apple does not have access; Apple will not use user data to develop and train AI; And Apple lives in a world where AI will be table issues in all software or hardware in the future.

How can Apple Intelligence break the AI ​​loop

What options are available for Apple?

It is too late to do something radical, such as deleting your efforts to implement AI and use its marketing power to highlight problems such as the need to collect training data, the personal information that should be exchanged, the environmental result of Hige and the potential inaccuracy of the results. It could have been a big sale two or three years ago, potentially modifying discussions on the AI ​​of smartphones.

Make this movement now, and it would be struck off as “you cannot do it, so you try to change the conversation; It will not work.

Tim Cook and his team could work on a different approach, putting AI in the handset with smaller LLM. Execution exclusively locally guaranteed that the data remains private and a smaller environment should be easier to code. It is a model that many Android smartphones use for the benefit. But these Android devices will quickly turn to an army of servers in the sky to access the cloud -based AI models for more precise results.

The final response would be to outsource AI efforts and work to provide the same AI advantages that Android can offer on mobile and chromeos and windows on more traditional computers. He started this path, Work with Openai To increase Siri’s results, but the answers so far do not feel like Apple’s typical standard.

What is happening for Apple Intelligence then?

Update, Sunday March 16, 2025: Write for Power on the newsletterMark Gurman describes the immediate steps that Tim Cook can take. Apple’s annual “Getaway” retirement for its prolonged management team will undoubtedly be dominated by discussions on what has not worked and what measures should be taken next. Although a large part of this is technical in nature, there must be discussions on how Apple’s IA efforts have been described to the public.

Apple Intelligence’s deliverables contrast with promises made not only to WWDC but also at the launch of the iPhone 16 family and current marketing on television and online.

By examining Apple’s senior management, Gurmn thinks that there will be no change of guard in the departments concerned or more in the organizational table:

“An upheaval would also suggest that the company has taken one of the pre -eminent names in the world in AI – Giannandrea – and has failed to put it in order to succeed, making another failed external rental to cook.

Tim Cook’s team, who has made the decisions about the implementation of Apple Intelligence, will have to be the same team that operates the way out of the situation.

Apple Intelligence harvests the seeds of yesteryear

Apple’s decision to focus on privacy has missed it poorly adapted to the rise of AI. The path they have taken in the ecosystem differs from competition and the divergence increases.

In simple terms, Apple must do this Big Data thing without data. If you believe the legends, Apple can deliver AI in a way that only Apple can. Frankly, if he can do the great thing with nothing, he will have won this legendary status.

Now, learn more about the Apple material improvements has been rebounded in To support the Apple Intelligence …