Apple is in talks with Anthropic and OpenAI to power the revamped version of Siri, potentially sidelining its own in-house AI models in the process. Here are the details.
According to Bloomberg’s Mark Gurman, Apple has asked both OpenAI and Anthropic to train customized versions of their large language models that could run on Apple’s Private Cloud Compute infrastructure.
Rather than relying on third-party cloud platforms like AWS or Azure, these models would live on servers powered by Apple silicon, giving the company more control over privacy.
Negotiations are ongoing, and no final decision has been made
Internally, the Siri revamp project has been led by Mike Rockwell and Craig Federighi, who were put in charge after Apple CEO Tim Cook reportedly “lost confidence” in executive John Giannandrea to lead the initiative.
Per Gurman’s report, the roadmap for a fully Apple-powered “LLM Siri” in 2026 is still technically active, but multiple rounds of testing led to the conclusion that Anthropic’s technology is currently better suited for Siri’s needs than what Apple has built so far. Apple even dispatched corporate development VP Adrian Perica to lead discussions with Anthropic.
Top comment by JR
This can't be surprising that AI has been a dumpster fire for Apple. One of Apple's largest selling points for years has been privacy. Unfortunately, AI is best when it has access to large amounts of data, its basically the opposite of privacy. Making AI that works for users on a privacy-focused platform is just going to hit road block after road block and will take a long time to get right because its a fundamentally different approach to AI.
AI is so huge in the tech industry that there's no way Apple's C-suite or their shareholders are going to be patient for them to figure it out in a meaningful way. What we've gotten so far are toys that do everything worse than what the other non-Apple tools do.
Here’s Gurman:
Apple’s investigation into third-party models is at an early stage, and the company hasn’t made a final decision on using them, the people said. A competing project internally dubbed LLM Siri that uses in-house models remains in active development.
Compensation is still an issue, as is internal morale
That said, Apple and Anthropic are reportedly far apart on pricing. Here’s Gurman:
While discussing a potential arrangement, Apple and Anthropic have disagreed over preliminary financial terms, according to the people. The AI startup is seeking a multibillion-dollar annual fee that increases sharply each year. The struggle to reach a deal has left Apple contemplating working with OpenAI or others if it moves forward with the third-party plan, they said.
Meanwhile, behind the scenes Gurman says that the potential shift to outside models isn’t sitting well with everyone at Apple. Morale has reportedly taken a hit among the Foundation Models team, with some team members feeling scapegoated for the company’s slow progress in generative AI, while others departed.
Apple is also facing a compensation problem, as companies like Meta and OpenAI have reportedly been paying packages as high as $40 million per year, far more than Apple typically pays for similar roles. In one recent case, the team behind Apple’s open-source MLX framework nearly quit en masse, before the company made counteroffers to keep them.
Best accessory deals on Amazon
- Anker 25,000mAh power bank, 100W, USB-C: $119.99
- Anker Foldable 3-in-1 wireless charger, 15W, Qi2: 20% off, at $71.99
- AirPods Pro 2: 20% off, at $199.00
- AirTag, 4 Pack: 24% off, at $74.99
- HomeKit Smart Plug, 15A, 4 Pack: 20% off, at $34.99
FTC: We use income earning auto affiliate links. More.
Comments