The call center is no longer a safe moat. Starlink is resolving more than 15,000 inbound customer support and sales calls every single day using xAI's Grok Voice. Not routing calls. Not transcribing them. Resolving them. Diagnosing hardware problems, shipping replacement equipment, and fulfilling over 3,000 orders per week. All without a human picking up. The model doing it is called grok-voice-think-fast-2.0, and while that name sounds like something a teenager named their gaming PC, what it's doing is genuinely significant.
Here's why this matters beyond Starlink's scale. Until now, the honest answer to "can AI handle my inbound calls?" was "kind of, for simple stuff." This changes that answer. If a system can troubleshoot satellite hardware over the phone and then ship a replacement without human review, the gap between enterprise-grade AI voice and what a small HVAC company or logistics shop could deploy is closing fast. If you run a call-heavy business, you now have a concrete benchmark to hold vendors against. Ask them: can your system resolve, not just deflect? That's the question.
You don't need a developer to build a business app anymore. Taskade Genesis lets you describe a process in plain English and get back a working app with AI agents, a database, and automations connected together. The newest addition is the part I find most useful: you can watch every automation step run in real time. That matters because most no-code tools are black boxes. Something breaks, and you're left guessing which step failed and why. With live execution visibility, you can actually debug it yourself, or at least tell someone where to look. Bookings, invoicing, content publishing pipelines. If you've been putting off automating something because you figured you'd need to hire someone, this is worth thirty minutes of your time. Free tier available.
Trending on GitHub
openai/skills: OpenAI's catalog of pre-built skills for Codex, their coding AI. Think of it as a recipe book of tasks you can hand to an AI developer. (46 stars today)
llvm/llvm-project: A foundational set of tools that help turn code into software that actually runs on computers. Not something you'd use directly, but it underpins most modern development toolchains. (23 stars today)
More Headlines
Miro is hiring founders internally to stay competitive in the AI era. About 40 entrepreneurs brought on in two and a half years, specifically for their owner mindset. Worth noting: that instinct is already your competitive advantage if you're running your own shop.
New research from Stax shows banks are pushing AI tools to small businesses that don't match what owners actually need. Cash flow visibility and payment friction are the real priorities. If your bank is pitching you AI features, ask whether any of them touch those two things.
BizTech lays out how AI adoption is exposing dirty data at small businesses. Scattered spreadsheets, disconnected systems, incomplete records. If your AI results feel inconsistent, that's probably why. Cleaning your data isn't glamorous, but it's the actual work.
MSN runs through how SMBs turn AI into lasting business value, with practical framing on integrating AI into workflows instead of running one-off experiments that go nowhere.
Analytics Insight rounds up 7 AI tools that could run a solo business in 2027. A decent list if you're a solopreneur thinking ahead.
Harvard has a new $699 entrepreneur bootcamp. Light on AI specifics, but worth a look if you're evaluating structured learning options.
-Scott
SmartOwner is published (almost) daily by the team at DigitalTreehouse. Want AI consulting or automations for your business? Reply to this email.
(Sources unavailable this run: Product Hunt.)