Apple's Strategies for Protecting Your Artificial Intelligence Data
It's no secret that Apple is working on AI features that will be launched with iOS 18 and macOS 15. When you update your iPhone, iPad, or Mac this year, you might experience a more natural-sounding Siri, or be able to generate emojis based on your conversation in Messages. These updates are exciting, but how is Apple going to safeguard your information while the AI processes these fancy new features?
While some reports claim that Apple will be running many of these features on-device, especially with its newer products, other rumors suggest the company plans on outsourcing much of the processing to the cloud. This is normal in the industry, as most AI processing happens in the cloud currently due to its intensive nature. Companies are constantly trying to enhance their NPUs (Neural Processing Units), which are specialized processors that deal exclusively with AI functions. Apple has used NPUs for years, but made a big fuss about the new M4 chip's powerful NPUs earlier this year, while Microsoft introduced a new AI-PC standard with its Copilot+ PC line.
On-device AI provides a more secure experience
No matter if your AI features run on your phone or in the cloud, you probably won't care as long as they function properly. However, running these features on-device has an inherently more secure experience. By pushing the processing to the cloud, companies put user data at risk, especially when the service processing the data needs to decrypt it first. The potential leaks include both the employees of the company and malicious hackers trying to breach the company's cloud servers.
This is already a significant problem with services like ChatGPT, and that's why I suggest not sharing personal information with most cloud-based AI services: Your conversations are not private, and are all being sent to these servers, both for storage and to train the AI model. Companies that prioritize user privacy, like Apple, prefer to use on-device solutions wherever possible, ensuring that user data remains isolated from others.
Utilizing 'Secure Enclave' to safeguard AI data
Even if Apple's hardware can handle the AI features they're developing, for older devices or features requiring too much power, the company may need to turn to cloud-based servers to offer those capabilities. But if a report from The Information, as mentioned by Android Authority, is correct, Apple may have found a solution: the Secure Enclave.
The Secure Enclave is already incorporated into most Apple devices in use today. It's part of the SoC (System on a Chip) that operates independently from the processor, and its primary function is to store your most sensitive information, such as your encryption keys and biometric data. This way, if the main processor is ever infected, the Secure Enclave keeps the data inaccessible to bad actors.
According to The Information, Apple is developing an AI-cloud solution that would send all AI user data to the Secure Enclaves of M2 Ultra and M4 Macs in its data centers. The server Macs in these facilities could process the request while maintaining encryption, and then send the results back to the user. This process would guarantee the safety of user data while allowing older devices to utilize Apple's latest AI features.
We still don't know if this is Apple's strategy until they announce it at WWDC, possibly. If Apple is tight-lipped about how they will secure AI user data, we may never find out. However, considering Apple positions itself as a company committed to user privacy, a solution that ensures encrypted cloud-based data would make a lot of sense.
Read also:
Despite Apple's plans to utilize cloud servers for some AI processing, the company is reportedly exploring a solution to safeguard user data by sending it to Secure Enclaves in their data centers. This method, which maintains encryption throughout the process, would allow older devices to utilize AI features while ensuring user data privacy, in line with Apple's commitment to preserving user privacy.
As Apple continues to develop its AI capabilities, the tech giant remains committed to prioritizing user data privacy, and this approach aligns with their efforts to enhance AI processing on-device while minimizing the risk of data leaks in the cloud. This commitment to AI data privacy is particularly notable, given the potential vulnerabilities associated with cloud-based AI services.
