The bigger story isn't that Apple's assistant got smarter, it's that a finally competent Siri is now the default on hundreds of millions of devices, and every other phone assistant has to answer to that.
For most of its life, Siri was the thing you apologized to after it ignored you. That arc, a decade-plus of jokes, workarounds, and gritted-teeth requests to "just open the timer app," is the backdrop for the verdict in this week's Vergecast from The Verge: "Siri is good now??" hosts David Pierce and Nilay Patel ask, before walking through their early hands-on impressions of Apple's new Siri AI release, dated June 12, 2026.
The surprise in the framing is doing structural work. Pierce and Patel are not declaring victory over frontier chatbots, and they are not claiming Apple has leapfrogged the field. Their point is more consequential than a hot take: a built-in assistant has crossed the threshold where most people can rely on it for most things, on the device they already carry, without opening a separate app.
That changes the baseline. For years, the AI assistant on your phone was something you tried, failed with, and stopped using. The default action was to open a chatbot tab or type a search. When the bundled assistant becomes competent, the path of least resistance flips. People stop reaching past their home screen for help, which means AI stops being a destination and starts being a layer underneath everything else.
This is where the category-level stakes show up. A "good enough" Siri on every supported iPhone is roughly a quarter of the active smartphone base in the United States and a larger share in several overseas markets, and the Vergecast hosts frame that footprint as the actual news, not a benchmark number. Alexa, Google Assistant, and the assistant layer on Android now have a moving target they cannot ignore, because the comparison is no longer to a chatbot you have to install. The comparison is to the thing that already answers when you talk to your phone.
The hosts are also careful about the ceiling. They are not arguing this Siri beats the best standalone assistants at the hardest tasks. The upgrade is iterative rather than bleeding-edge, and that honesty is part of the constructive case. A floor-raiser is not a frontier model. A floor-raiser shipped by default, though, is the more important product event for most users, because the median experience is what shapes expectation, and expectation is what every other assistant is now measured against.
The user-habit angle is where the story actually moves. When the on-device helper handles "set a reminder for when I get home," "draft a reply to this message," "summarize this article," and "find the photo of the dog from last summer" without flinching, the surface area for AI in daily life quietly expands. The new behavior is not "I tried an AI thing." It is "I asked my phone." That shift is harder to see than a benchmark jump, and it travels further, because every successful interaction trains the user to skip the extra step next time.
There is a real critique baked into the decade of justified frustration that preceded this release, and the Vergecast is not asking anyone to forget it. Siri earned the jokes. The structural point is that the assistant category has been waiting for one of the default-bundled options to clear the bar, and the consequence of that is not just a better Siri. It is a different relationship between users and the device in their pocket, and a more uncomfortable one for every other assistant still trading on being the alternative people open when the built-in one fails.
What to watch next is whether the early impression holds under sustained use, whether Apple can keep latency and reliability consistent across the long tail of requests, and whether competing platforms respond with their own default-bundled upgrades or keep asking users to install a separate app for the same job. The Vergecast's question mark is doing real work here. It is the first time in a long time that question has been asked in good faith, and the answer is the kind of product event that resets the floor for the entire category, not just the one assistant that finally cleared it.