Apple has committed to the White House’s voluntary guidelines for creating safe,secure,and trustworthy AI.According to a press release on Friday,Apple will soon introduce its generative AI product,Apple Intelligence,to its core offerings,making generative AI accessible to Apple’s 2 billion users.

　　Apple joins 15 other tech companies—including Amazon,Anthropic,Google,Inflection,Meta,Microsoft,and OpenAI—in agreeing to the White House’s guidelines for generative AI development set in July 2023.Although Apple had not detailed its AI integration plans for iOS at that time,its announcements at WWDC in June made clear its commitment to generative AI,including a partnership to embed ChatGPT in the iPhone.By aligning with the White House’s guidelines,Apple may be attempting to demonstrate its compliance ahead of potential regulatory scrutiny.

　　While Apple’s commitment to the White House’s guidelines is voluntary and not legally binding,it represents a preliminary step towards ensuring AI safety,security,and trustworthiness.This move follows President Biden’s AI executive order in October,with additional AI regulatory bills progressing through federal and state legislatures.

　　The White House’s guidelines for AI companies include:

　　1.Red-Teaming AI Models:Conducting adversarial testing on AI models before public release and sharing results with the public.

　　2.Confidential Handling of AI Model Weights:Ensuring AI model weights are treated confidentially and worked on in secure environments with limited access.

　　3.Developing Content Labeling Systems:Implementing systems like watermarking to help users distinguish AI-generated content.

　　The Department of Commerce is also set to release a report on the benefits,risks,and implications of open-source foundation models.Open-source AI has become a contentious issue,with some advocating for restrictions on model weight accessibility for safety reasons,while others argue that such limitations could stifle AI innovation and research.The White House’s stance on this issue could significantly influence the broader AI industry.

　　Moreover,federal agencies have made considerable progress on AI-related initiatives since the October executive order.These include more than 200 AI-related hires,granting over 80 research teams access to computational resources,and releasing various frameworks for AI development.