One day I visited a customer that was usually super casual, but everybody was wearing a tie.
Turns out there was one guy who always wore one. The one day he wanted to skip it, he asked management for permission. They said no and somehow that ended with the whole company order to dress like they were going to a wedding every day.
I keep thinking about that story whenever I see people trying to solve unlikely AI risks by adding more and more restrictions. Sometimes the workaround becomes more absurd than the original problem.
@steipete Being a remote worker doesn’t mean you don’t need an office. I’ve been remote working for 25+ years now, got my dedicated office (15 minutes drive from home) after the first 3 years. It’s my temple.
@spyced In Spanish, we say “Si no la gana, la empata” (“If they can’t win it, they tie it”) about people who always have an excuse or comeback when they’re wrong and never admit it.
Me llama la atención que nadie piense lo más obvio. El presidente a 8 días de asumir debía estar a mil. La automotora, por tenerlo de cliente, le ofrece un descuento/atención a través de algún baraja de su entorno. La diferencia le sirve, da el OK verbal. Le mandan la camioneta. Ni ve la factura. La automotora sub factura para que la atención le quede gratis bajando el IVA.
Todo el resto es un zafarrancho tremendo de su equipo de asesores que jamás hizo una permuta.
@graemelamperson We have thousands of cons, but there’s a good thing about most Uruguayans, we really don’t care who you are. You can be Elon eating some panchos at La Pasiva, nobody will even stare at you. And you don’t have priority, because here “naides es más que naides”.
Joined a banking AI roundtable with 25+ Heads of AI from some of the world’s largest banks. Many oversee tens of thousands of Copilot seats. Only one had even heard of @steipete’s @OpenClaw.
I understand the regulatory constraints, but the disconnect between AI leadership and hands-on experimentation is larger than I expected.
Nobody knows where this is going, but I suspect AI usage costs will become part of a programmer’s personal toolkit—like buying books, hardware, or investing in education.
Out of 5,000 developers, maybe 500 will emerge dramatically more productive. Those people won’t just keep their jobs—they’ll redefine them, and they’ll be compensated accordingly.
We’re living through a difficult transition for our industry. Think of money spent on AI tokens as an investment in your future earning power.
Uber’s COO has said that it’s getting “harder to justify” its AI costs because there was no way to show a link between AI spend and any meaningful increase in useful features. This is the first time I’ve seen a company say this directly.
https://t.co/xUhZvtpwah
No, finite state machines are NOT equivalent to Turing machines. There's a whole hierarchy of what an automata can generate (just google Chomsky hierarchy). Any decent CS course will teach you about it, pick any book about automata theory, or even ask your favorite LLM.