Apple is considering enlisting the help of OpenAI or Anthropic to power its AI-upgraded Siri, according to a report from Bloomberg’s Mark Gurman. As Apple continues to struggle with the development of an upgraded “LLM Siri,” it reportedly asked OpenAI and Anthropic to create versions of their large-language models to test on the company’s private cloud infrastructure.

As reported by Bloomberg, Rockwell asked his team to test whether Anthropic’s Claude, OpenAI’s ChatGPT, or Google’s Gemini performs better at handling basic requests compared to its own models, with Anthropic’s apparently seen as the most promising. While Google has Gemini AI features for Android and its Pixel lineup, Samsung licenses Google’s AI model for its phones. It is also reportedly close to cutting a deal with Perplexity, which already has a tie-up with Motorola. Earlier this month, Bloomberg reported Apple executives had considered acquiring Perplexity to help boost its AI ambitions.

LLM Siri was largely absent from Apple’s Worldwide Developers Conference earlier this month, where SVP of worldwide marketing Greg Joswiak admitted that the technology “didn’t hit our quality standard.” Bloomberg notes that Apple’s plans to incorporate a third-party AI model into Siri are still at an “early stage” and that it’s still considering using in-house models.