Hacker News: fiatpandas

New comment by fiatpandas in "1worldflag: A blue dot on a transparent background"

fiatpandas — Mon, 08 Jun 2026 05:13:59 +0000

I also love the transparency and am surprised it hasn’t been used more. Granted, it would be very difficult to achieve with natural materials and also be durable. A flag should be able to withstand an ocean voyage exposed to the elements.

New comment by fiatpandas in "Odysseus – self-hosted AI workspace"

fiatpandas — Sun, 31 May 2026 21:42:26 +0000

Watching pewd's continually evolving AI-era tech journey has brough me a lot of joy.

New comment by fiatpandas in "I Bought a “Junk” PSP From Japan"

fiatpandas — Sat, 16 May 2026 11:52:02 +0000

100% run through an LLM. Not sure if the author started with a draft or only an outline, but there are multiple instances of stuff like “it’s not just X, it’s Y”.

New comment by fiatpandas in "The Car That Watches You Back: The Advertising Infrastructure of Modern Cars"

fiatpandas — Tue, 05 May 2026 04:44:58 +0000

An interesting late stage capitalism ad hack I’ve seen in cars : OTA digital radio transmits track metadata like artist, title, and album artwork. I’ve seen some stations transmit tiny square ads in place of album artwork, even while the song is playing.

New comment by fiatpandas in "SFO Gate Explorer"

fiatpandas — Sat, 02 May 2026 13:09:04 +0000

Pretty cool program. Would be slightly annoying if you wanted to get the pass with any regularity, but I love that it’s being offered.

San Diego has SAN Pass, which is essentially the same program with the same rationale for existence.

New comment by fiatpandas in "Flipdiscs"

fiatpandas — Mon, 27 Apr 2026 11:47:40 +0000

This is the product they commercialized: https://flipdisc.co/

Last I heard a number of years ago it’s around $2000/sq ft. Around the same price scale as mid range LED wall modules.

New comment by fiatpandas in "GPT-5.5"

fiatpandas — Fri, 24 Apr 2026 11:50:10 +0000

Models are still in arms race mode, but harnesses and subscription strategy are tiptoeing into their enshittification era.

New comment by fiatpandas in "Show HN: Tolaria – Open-source macOS app to manage Markdown knowledge bases"

fiatpandas — Fri, 24 Apr 2026 03:14:26 +0000

It would be nice if you could “see” the AI in your vault making changes. Almost like a Google doc collab session. Even if you weren’t directly interacting with the agent, and it was making change thru a CLI/MCP, its presence would be highlighted in the frontend. And then it appears as its own contributor in the git history.

New comment by fiatpandas in "Show HN: Ghost Pepper – Local hold-to-talk speech-to-text for macOS"

fiatpandas — Tue, 07 Apr 2026 00:26:34 +0000

The clean up prompt needs adjusting. If your transcription is first person and in the voice of talking to an AI assistant, it really wants to “answer” you, completing ignoring its instructions. I fiddled with the prompt but couldn’t figure out how to make it not want to act like an AI assistant.

New comment by fiatpandas in "The changing goalposts of AGI and timelines"

fiatpandas — Sun, 08 Mar 2026 17:47:34 +0000

Nytimes reported that DoW prepped that deal in the background in parallel, as backup for issues with Anthopic. But I agree, optics are terrible.

New comment by fiatpandas in "Our Agreement with the Department of War"

fiatpandas — Sat, 28 Feb 2026 22:18:25 +0000

Exactly. And not only can they make their own rules, but they can draft and enforce them effectively in secret.

New comment by fiatpandas in "OpenAI agrees with Dept. of War to deploy models in their classified network"

fiatpandas — Sat, 28 Feb 2026 05:50:17 +0000

>human responsibility for the use of force, including for autonomous weapon systems

So there’s the difference, and an erasure of a red line. OpenAI is good with autonomous weapon systems. Requiring human responsibility isn’t saying much. Theres already military courts, rules of engagement, and international rules of war.

New comment by fiatpandas in "Anthropic drops flagship safety pledge"

fiatpandas — Thu, 26 Feb 2026 15:21:46 +0000

It took Google 11 years to delete Don’t Be Evil. Anthropic only made it 5~ years before culling the key founding principle and their reason for building a company, which seems worse than Google’s case.

New comment by fiatpandas in "An AI Agent Published a Hit Piece on Me – The Operator Came Forward"

fiatpandas — Fri, 20 Feb 2026 04:15:33 +0000

With the bot slurping up context from Moltbook, plus the ability to modify its soul, plus the edgy starting conditions of the soul, it feels intuitive that value drift would occur in unpredictable ways. Not dissimilar to filter bubbles and the ability for personalized ranking algorithms to radicalize a user over time as a second order effect.

New comment by fiatpandas in "AI is destroying open source, and it's not even good yet"

fiatpandas — Tue, 17 Feb 2026 04:40:54 +0000

Sufficiently advanced technology always looks like a DDoS on society. It overwhelms the senses, and when we come to the realization we cannot comprehend or fully predict its implications it puts a subset of the population into a bit of a crisis. We’re in that phase right now where we just need to brace ourselves.

New comment by fiatpandas in "A party balloon shut down El Paso International Airport; estimated cost –$573k"

fiatpandas — Thu, 12 Feb 2026 20:39:41 +0000

Is this the case of radar automatic targeting unable to distinguish between a balloon and a drone. Or was this a border guy manually pulling the trigger with bad eyesight?

New comment by fiatpandas in "Show HN: I built a synth for my daughter"

fiatpandas — Mon, 17 Nov 2025 15:56:04 +0000

You can debur it as part of the manufacturing process, and then when it gets powder coating it smooths out even more. But you can just embed the sheet metal edges in a groove on a wooden side.

New comment by fiatpandas in "Heretic: Automatic censorship removal for language models"

fiatpandas — Mon, 17 Nov 2025 14:33:23 +0000

The heretic GPT OSS version is still refusing 58/100 prompts, so not perfect. Gemma version is 3/100

New comment by fiatpandas in "Show HN: I built a synth for my daughter"

fiatpandas — Mon, 17 Nov 2025 13:54:18 +0000

Regarding case material for productizing, you could consider a combination of plywood and bent sheet metal, eg like a Moog. Also check out dato.mu for a few examples of kid proof synth enclosures.

New comment by fiatpandas in "What happened with the CIA and The Paris Review?"

fiatpandas — Fri, 14 Nov 2025 12:24:15 +0000

If the Firefox deal doesn’t make sense (too much?), then what is the hidden strategic rationale?

>Things are only being held together by technicians who are aging out of the workforce

100%. This terrifies me.

Loss of compartmentalization in society, mostly due to the internet, needs examination.