My AI agent booked this hair appointment


I'm writing this from a salon chair on Saturday night; to be sent on Monday morning.

I had an event Saturday evening and needed to do something about my hair.

I didn't book the appointment. My AI agent did.

I use two of them, Muse and Grokbot, depending on what I'm trying to get done. This morning I needed a same-day blowout. The agent checked availability across four salons, matched the open slots against my calendar, and I ended up here with Karina.

Sitting here, I scrolled back through everything I've asked these agents to do over the past few weeks. The list surprised me.

Travel and logistics

  • Monitoring flights
  • Building a full conference trip: itinerary pulled from Gmail, a panel prep brief, the agenda, leave-by times, Uber coordination
  • Weather heads-ups for a homecoming storm, the Oct 17 game, and the week the roofers are coming

Appointments

  • Researching two salons for cut and color, then figuring out how each one actually books (one through Mangomint, one by phone only)
  • Checking my massage therapist's live Square availability, booking, attempting reschedules, and drafting a text to a backup therapist
  • Reminders with real timing logic: when to order an Uber to DCA, when to head out of a conference, when a massage cancellation window closes, when SAT scores drop

Kids

  • SAT registrations, Bluebook setup, and a test-center decision chain, including a correction on the $34 fee
  • Water polo camp registration, practice schedules, tournament travel, a school service day
  • Tracking down the school portrait company's customer care email and drafting the follow-up for missing photos
  • A daily 9 AM scan of eight college Instagram accounts for new men's water polo commits, reporting only the new ones with position, handedness, high school, class, and town

Money and inbox

  • Flagging a check-deposit relay and an unrecognized Venmo request
  • Auditing my personal Gmail for cancellable subscriptions, with renewal dates and cancel steps. It also flagged a correction: the two Claude charges belonged to two different people's accounts.
  • Tracking roof payments and schedule

Work

  • Fixing my calendar: a game on the wrong day, a duplicate lunch, a time block missing from one of my two calendars
  • Spotting things I still owe clients, like next steps for one and a Shopify setup guide for another

Most of this is the connective tissue of a life. The checking, cross-referencing, and remembering that usually lives in my head at 11 PM.

The most valuable moments were the catches. The $34 fee. The two Claude accounts. The salon that only takes phone calls. Noticing what's off is where the time actually comes back.

For the leaders I work with, this is the useful mental shift. Chatbots answer questions. Agents carry errands across apps (email, calendar, booking systems, the web) and come back with a result. If you've only used AI to draft emails, try handing it one errand this week that touches two or three systems. A booking. A reschedule. A subscription audit.

Then watch what it catches.

Alex

AZAI Advisory

As an AI Coach, Advisor, and Agent Builder, I help organizations and business leaders harness the power of artificial intelligence to boost productivity and streamline operations. I enable organizations to navigate the transformative landscape of AI, educating teams, identifying operational and strategic opportunities with AI and creating a framework for safe and transparent use of data in the organization.

Read more from AZAI Advisory

Nate B. Jones wrote something last week that I've been chewing on. He described research (from Epoch and METR) where the same Claude model rebuilt a roughly 16,000-line scientific toolkit from its behavior alone, without seeing the code. Then, asked to customize a smaller library, it produced something unusable. Same model. Wildly different results. His explanation: AI improves fastest where results are cheap to check, like proofs and tests. It improves slowest where feedback is slow, like...

I recently spent a session with an investor walking through everything Claude can do. At the end, he asked: "Okay, now how do I get all the really good investment ideas from Claude?" A fair question. Also the wrong one. Around the same time, I was working with a hedge fund team on the same material. Same tools, same techniques. They went somewhere completely different with it. They connected the dots on their own. They took what I taught them and ran their own view of what makes a good...

Last week I sat down with the COO and general counsel of a hedge fund that's about to start AI training. We were talking governance, and I opened with a picture. Right now, every analyst who uses Claude describes the firm in their own words. How you invest, what you care about, what a good idea looks like. Twenty-two people, twenty-two descriptions. But your fundraising decks present a different picture: One firm, one description of the company and its investment strategy. Signed off by the...