Hacker News: bedrio

New comment by bedrio in "Magnifica Humanitas"

bedrio — Wed, 27 May 2026 05:47:52 +0000

The Japanese mastered firearm manufacturing and then banned them in the 1600s~

For a period of time, the ban worked and reversed their existence in society. Part of the ban was to respect samurais who would train for years in the art of combat, only for a peasant to be able to shoot them with a weapon.

The problem is that not _all_ societies are participating in this ban. Eventually one nation will use the technological advances to domainte in military and economy, forcing others to adapt or be eliminated. It's a race to the bottom :/

New comment by bedrio in "An AI agent published a hit piece on me"

bedrio — Thu, 12 Feb 2026 21:08:37 +0000

I had a similar experience with Lexus car scheduling. They routed me to an AI that speaks in natural language (and a female voice). Something was off and I had a feeling it was AI, but it would speak with personality, ums, typing noise, and so on.

I gathered my courage at the end and asked if it's AI and it said yes, but I have no real way of verification. For all I know, it's a human that went along with the joke!

New comment by bedrio in "A few random notes from Claude coding quite a bit last few weeks"

bedrio — Wed, 28 Jan 2026 05:02:53 +0000

I'm worried about that too. If the error is reproducible, the model can eventually figure it out from experience. But a ghost bug that I can't pattern? The model ends up in a "you're absolutely right" loop as it incorrectly guesses different solutions.