Apple Joins Voluntary AI Safety Initiative

Apple has joined a voluntary commitment proposed by the White House to develop safe and ethical artificial intelligence, becoming the 16th technology company to support the initiative. This decision was made on the eve of the launch of Apple Intelligence, its own generative AI model, which will cover more than 2 billion Apple users worldwide.

Image Source: Apple

Apple joins 15 other tech giants, including Amazon, Google, Microsoft and OpenAI, that have pledged to follow White House principles on AI development in July 2023, TechCrunch reports. Notably, Apple didn’t reveal its plans to integrate AI into iOS until the recent WWDC conference in June, where the company announced its intentions to develop generative AI, starting with a partnership with ChatGPT in the iPhone.

Analysts see Apple’s decision as an attempt by the company, known for its controversial relationship with regulators, to demonstrate its readiness to cooperate in the field of AI. Amid growing pressure from lawmakers and public concerns about the unchecked development of AI, Apple has decided to show itself as a responsible company willing to follow ethical principles.

As part of the voluntary commitment, companies commit to thoroughly testing the safety of AI models before releasing them to the public, and to making the test results available to the public. In addition, they must ensure the confidentiality of the AI ​​models they develop by limiting access to the development to a wide range of employees. An agreement has also been signed on a system for labeling AI-generated content so that users can easily distinguish it from human-generated content.

While these commitments are voluntary, the White House views them as “the first step toward creating safe and trustworthy AI.” In addition, a number of bills to regulate it are currently being considered at the federal and state levels.

In parallel, the US Department of Commerce is preparing a report on the potential benefits, risks and consequences of open source AI models. At the same time, closed-access AI models have become the subject of heated debate, as limiting access to powerful generative models can negatively impact the development of startups and research in the field of AI.

US authorities also noted significant progress by federal agencies in meeting the goals set by the October order. To date, more than 200 AI specialists have been hired, more than 80 research groups have access to computing resources, and several frameworks for AI development have been released.