Hey there! If you’ve been keeping an eye on Apple’s AI developments, you might have heard about some of the challenges they’re facing with Siri’s next-gen features. It turns out that scaling these advanced AI capabilities isn’t as straightforward as it might seem, especially when you’re trying to balance rapid rollout with maintaining top-notch quality across a vast range of devices.
According to Bloomberg, Apple’s internal tests have shown that the new Siri functions are hitting accuracy rates between 66 and 80 percent. Robby Walker, who’s a Senior Director at Apple, shared these insights during a big meeting with the Siri team. Initially, these Siri upgrades were supposed to roll out this spring, offering features like managing personal data, handling emails, and finding specific images. But now, it looks like we’ll have to wait until iOS 19 in 2026, and even that date isn’t set in stone.
Walker didn’t hold back in describing these delays as “ugly and embarrassing,” especially given Apple’s earlier marketing push. At the WWDC developer conference in June 2024, they only managed to show a basic prototype, which was a bit of a letdown compared to the earlier buzz.
To pick up the pace, Apple plans to have more senior managers report directly to AI chief John Giannandrea. While no immediate leadership changes are expected, the goal is to make Siri more conversational and competitive with other AI chatbots by 2027. Despite these setbacks, Walker praised his team’s dedication to quality over speed, which is a tough balance to strike.
Now, you might wonder, why is an 80 percent success rate a big deal? Well, given the nature of generative AI systems, this might actually reflect the current technological limits. Even with a 95% success rate, millions of errors could still occur across Apple’s massive user base. That’s a huge challenge, especially when compared to companies like OpenAI, which can launch experimental features as “research previews” with known risks.
Google is facing similar issues with its AI search offerings. While smaller players like OpenAI and Perplexity can afford to release less-than-perfect AI search products, larger companies like Google and Apple are under more scrutiny. Google’s started to manage expectations by acknowledging that early-stage AI products won’t always be perfect.
Other tech giants, including Meta and Microsoft, are also adjusting their strategies, openly recognizing errors in their AI products. It’s an interesting time for tech, with the current U.S. political climate allowing companies to release imperfect AI features with fewer repercussions.
So, while Apple’s heavily advertised Siri enhancements are postponed until at least 2026 due to only achieving 66 to 80 percent success in internal testing, it’s clear that they’re committed to maintaining high-quality standards. This commitment to quality means they won’t release incomplete AI technologies, unlike smaller competitors who can test their products more freely. It’s a challenging landscape, but one that Apple seems determined to navigate carefully.
