Apple is reportedly exploring partnerships with OpenAI or Anthropic for the development of an enhanced version of its virtual assistant, Siri. According to a report by Bloomberg’s Mark Gurman, the tech giant has requested that both companies create versions of their large language models for testing on Apple’s private cloud infrastructure.
In recent months, Apple has faced challenges in revitalizing its AI-driven Siri after postponing the rollout of the revamped assistant in March. Following this delay, CEO Tim Cook appointed Mike Rockwell, who heads the Vision Pro team, to lead the AI and Siri projects, replacing former chief John Giannandrea due to diminishing confidence in his leadership.
As noted by Bloomberg, Rockwell has instructed his team to evaluate the performance of Anthropic’s Claude, OpenAI’s ChatGPT, and Google’s Gemini in handling basic user requests. Initial assessments suggest that Anthropic’s model may have the edge. Meanwhile, Google’s Gemini AI is already integrated into Android devices and the Pixel range, and Samsung has partnered with Google to utilize its AI capabilities. Additionally, Samsung is reportedly close to finalizing a deal with Perplexity, a company already collaborating with Motorola for AI features. Recent reports indicate that Apple executives have also contemplated acquiring Perplexity to advance their AI initiatives.
Siri’s anticipated LLM integration was notably absent from Apple’s Worldwide Developers Conference earlier this month, where Greg Joswiak, SVP of worldwide marketing, acknowledged that the technology “didn’t hit our quality standard.” Bloomberg emphasizes that the discussions around incorporating a third-party AI model into Siri remain in the preliminary stages, with Apple still weighing the possibility of utilizing its internal models.
