August 3, 2026
TechCrunch's verdict on the new Siri overhaul is the right one: it's genuinely good, and that's exactly the problem. Apple spent years promising a Siri that could actually understand context and follow through on tasks, and it's finally delivered that. But the bar it's clearing was set for 2023. The rest of the market has moved from chatbots to agents that write code, generate media, and execute multi-step work without hand-holding. A voice assistant that finally answers questions correctly and sets reminders reliably is a real engineering achievement, but it's competing against products that stopped thinking of themselves as assistants a while ago.
For business buyers, the lesson isn't about Siri specifically -- it's about how quickly 'good enough' becomes 'too late' in this market. Apple had every advantage: hundreds of millions of devices, deep pockets, and years of runway. It still couldn't ship on its original timeline, and by the time it did, the definition of a useful AI assistant had moved past what it built. That's a caution for any company evaluating AI tools on roadmap promises rather than what's shipping today. If a vendor's pitch is 'wait for our next release, it'll be worth it,' Apple's Siri saga is the cautionary tale to keep in your back pocket.
While Apple was polishing Siri, TechCrunch reported that House spending records show ChatGPT is already the dominant paid AI tool on Capitol Hill, used across congressional offices to draft memos, summarize legislation, and handle constituent replies. There's no mystery to why: ChatGPT was available, worked well enough, and staffers adopted it before anyone drew up a formal AI strategy. That's the pattern showing up everywhere, not just in government -- the tool that gets embedded first tends to stay embedded, regardless of what launches later with more fanfare.
This is the real story buried under both headlines: habit formation beats feature parity. Once a workflow is built around a tool -- once staffers know the prompts that work and trust the output enough to send it out the door -- switching costs become real, even if a competitor ships something objectively better. That's true in a congressional office, and it's just as true inside a sales team running its pipeline through a CRM or an ops team that's standardized its workflow automation around one AI-assisted tool. The vendors who win aren't always the ones with the most impressive demo -- they're the ones who got there first and made themselves hard to rip out.
Put these two stories side by side and you get a genuinely useful warning. Apple shows what happens when you optimize for a polished, closed system and ship late. Congress shows what happens when an organization adopts a general-purpose tool out of convenience with no real evaluation process -- you end up standardized on whatever staffers happened to download first, whether or not it's the right long-term fit for handling sensitive legislative or constituent data. Neither path is obviously right. The honest takeaway is that the tools worth betting on are the ones flexible enough to absorb whichever AI model turns out to matter next, instead of locking you into one assistant's roadmap or one habit formed by default. Teams that build their own internal tools rather than betting everything on a single closed assistant put themselves in a much stronger position when the next model, the next Siri, or the next ChatGPT competitor inevitably shows up.
Which matters more to your organization right now: an assistant that's finally competent, or the fact that your team already has years of habit built around a completely different tool?
Sources