As Apple around the world Keynote of developers conferences ended on Monday, market observers could not fail to notice that the company’s share price has dropped, perhaps a reaction to the relatively low Apple approach to enabling AI compared to most of its competitors. Despite this, the Functions and improvements of Apple Intelligence were plentiful, and although some are powered by the company’s privacy and the cloud -oriented cloud platform known as a private cloud, many of them work locally on devices supporting Apple Intelligence.
The recent Apple message review function automatically transfers texts from phone numbers and accounts with which you never interact to the “broadcaster” folder before. This function automatically detects messages sensitive to time, such as login codes or updates to supply food and still supplies them to the main inbox, but also scans messages that seem to be fraud and place them in a separate spam folder. All this sorting is done locally using Apple Intelligence. Similarly, the extended connection review function will automatically and locally receive undefined phone calls, ask for details about the caller and write the answers so that you can decide if you want to pick up the call. Even live translation adds translation of the language in real time to connections and messages by means of local processing.
From a privacy perspective, local processing is a golden standard for AI. The data never leave the device, which means that there is no risk that it may end somewhere unintentional as a result of a cloud travel. And recent functions, such as spam and “sender crush”, sorting messages, testing calls for undisturbed phone numbers and live translation tools seemed to be designed with the privacy strategy as a difference in the already crowded AI field.
In addition to being amiable to privacy, local processing has different benefits, such as enabling the availability of services based on artificial intelligence and accelerating certain tasks, because the data does not have to be sent to the cloud, processed and then sent to the device. If AI functions are widely available and available, most companies are constrained, trying to include aged, low devices, which many of their clients probably exploit, which may not be able to support local artificial intelligence. However, Apple has a smaller need to turn on, because it manufactures both hardware and software and has already imposed restrictions that Apple Intelligence can only work at all in the latest devices models.
There are also other restrictions for Apple Intelligence, and the company offers opt-in integration with some generative services and other companies to expand functionality. For example, in the case of ChatGPT OPENNAI, users must enable integration, and Apple will then encourage the user to confirm every time it goes to the message of CHATGPT. In addition, users can decide to log in to the CHATGPT account, in which their inquiries will be subject to normal openai principles, or can exploit chatgpt without logging in. In this scenario, Apple claims that it does not connect Apple ID or other identifier with questions and concealed the IP addresses of users.
Apple has intensively invested to develop a private cloud to maintain powerful safety and privacy guarantees in the field of AI processing in the cloud. Other companies even began to create similar safe and sound cloud programs for products and services that particularly focus privacy as a key function. But the fact that Apple is still implementing local processing of recent functions, if possible, may indicate that privacy is not just an intellectual priority in the company’s approach to artificial intelligence, it may be a business strategy.