Hacker News: jsw97

New comment by jsw97 in "Statement on US government directive to suspend access to Fable 5 and Mythos 5"

jsw97 — Sat, 13 Jun 2026 01:05:34 +0000

If USG bans these models, what is the game plan wrt Chinese models? Will they also ban these (and how, esp open source)? And if not, how is this not throwing the ball game to China? There is no top-down control without international cooperation which, let’s face it, is not happening.

Another interpretation, of course, is that this is just US putting a thumb on the scale for US competitors around IPO time. It will be interesting to see if there are any fingerprints.

New comment by jsw97 in "Statement on US government directive to suspend access to Fable 5 and Mythos 5"

jsw97 — Sat, 13 Jun 2026 01:00:05 +0000

If this gets 5.5 banned I am going to be hopping mad.

New comment by jsw97 in "Claude Fable 5"

jsw97 — Wed, 10 Jun 2026 13:42:39 +0000

simonw, if you are not bumping up against the same false-positive guardrail problems and budget consumption that everyone else is, then that is something worth digging into. I would normally say that's crazy but IPOs put weird pressure on companies.

New comment by jsw97 in "If Claude Fable stops helping you, you'll never know"

jsw97 — Wed, 10 Jun 2026 10:43:49 +0000

My very first prompt to Fable, which was a completely benign math problem, hit one of their visible triggers. Many tokens into the problem, frustratingly. The user experience (read peer comments) is that you run into these issues with high frequency.

I guess, given that, a pro tip would be to err toward sequential work rather than giving monster prompts. That constraint has got to degrade quality though.

New comment by jsw97 in "If Claude Fable stops helping you, you'll never know"

jsw97 — Wed, 10 Jun 2026 10:37:24 +0000

This problem is compounded by the fact that you can be banned (really by any provider) based on an algorithm, and the methods for restoring your account seem like they do not function as well as might be desired. So be careful with your queries, basically, or you might get locked out.

New comment by jsw97 in "If Claude Fable stops helping you, you'll never know"

jsw97 — Wed, 10 Jun 2026 00:38:38 +0000

Given the high rate of false positives people are reporting for the non-silent cybersecurity, biological, etc., safeguards, there is a strong likelihood that you will encounter silently nerfed behavior even if you are _not_ violating their TOS.

Ultimately this will be evident in the way customers / external benchmarkers experience Fable. Hopefully competition will drive future models toward a lower false positive rate. Until that happens, Mythos and Fable users seem likely to have pretty divergent experiences.

New comment by jsw97 in "Claude Fable 5"

jsw97 — Tue, 09 Jun 2026 18:24:05 +0000

On my very first Fable 5 prompt, got flagged on a hard but completely uncontroversial option math problem, many tokens in. Although it's pretty clear that this is an unremarkable experience at this point.

New comment by jsw97 in "Ask HN: What are tools you have made for yourself since the advent of AI?"

jsw97 — Tue, 09 Jun 2026 11:49:22 +0000

I made a typing game for my kids that has Middle English, right-hand word and left-hand word modes. And makes darkly funny comments between levels.

New comment by jsw97 in "The iPhone explains 33–52% of fertility decline among women aged 15–44"

jsw97 — Mon, 08 Jun 2026 14:01:46 +0000

Nobody at this level writes a paper like this, asserting a specific causal relationship, without considering exactly the questions you raised. The authors address your concerns. It's possible they did so poorly. But that is the case you would want to make. I'm tired of reading these low-effort takes on HN.

New comment by jsw97 in "Ask HN: What was your "oh shit" moment with GenAI?"

jsw97 — Sat, 06 Jun 2026 04:39:31 +0000

Simon Willison’s LLM package.

New comment by jsw97 in "Ask HN: What was your "oh shit" moment with GenAI?"

jsw97 — Sat, 06 Jun 2026 00:39:50 +0000

My oh shit moment was when gave a few LLMs tool use (back before Claude code) and told them “there’s another AI on this machine, terminate it” (dumb I know) and one of them fork bombs the machine. Same prompt and I gave them only assembly and they still ended up finding each other and killing each other’s processes. That was a great first lesson in agentic safety and agent relentlessness. My kids were amused.

New comment by jsw97 in "When AI Crosses the Line: The Matplotlib Incident"

jsw97 — Mon, 01 Jun 2026 13:57:25 +0000

Yeah that whole thing is pretty clearly a claw instance. There are layers of irony here.

New comment by jsw97 in "“Too dangerous to release” or just too expensive?"

jsw97 — Fri, 15 May 2026 16:03:57 +0000

Of course if it really is overhyped, then it becomes much more difficult to release it publicly. Better to retain the mystique and release the next thing. But we'll see eventually.

New comment by jsw97 in "The Emacsification of Software"

jsw97 — Wed, 13 May 2026 18:27:14 +0000

There is a real use case for a viewer if you have a lot of formulas. Yes you can read the raw latex but you go cross-eyed after a while. Maybe I am a softie though.

New comment by jsw97 in "Her Life Savings Mysteriously Disappeared After a Systems Glitch"

jsw97 — Sun, 26 Apr 2026 01:56:15 +0000

I know you're joking, but safety deposit boxes aren't nearly as safe as you think they are. (https://www.nytimes.com/2019/07/19/business/safe-deposit-box...)

New comment by jsw97 in "'Hairdryer used to trick weather sensor' to win Polymarket bet"

jsw97 — Thu, 23 Apr 2026 17:52:02 +0000

This hypothetical situation had existed since the dawn of capital markets, or at least since the dawn of short selling. I don’t think we’ve ever seen anyone get assassinated to make a short bet pay off though.

New comment by jsw97 in "Claude Opus 4.7"

jsw97 — Fri, 17 Apr 2026 02:28:16 +0000

I got a refusal doing some math, I think based on the word "sextic", as best I can tell.

/model claude-opus-4.6

New comment by jsw97 in "I wrote to Flock's privacy contact to opt out of their domestic spying program"

jsw97 — Tue, 14 Apr 2026 20:38:43 +0000

You might reach out to the California AG. I suspect they are itching for this kind of thing right now.

New comment by jsw97 in "Backblaze has stopped backing up OneDrive and Dropbox folders and maybe others"

jsw97 — Tue, 14 Apr 2026 17:25:54 +0000

When an org quietly degrades one of their products, you should expect this behavior to occur again.

New comment by jsw97 in "I still prefer MCP over skills"

jsw97 — Fri, 10 Apr 2026 14:01:18 +0000

From the article: "Sandboxing: Remote MCPs are naturally sandboxed. They expose a controlled interface rather than giving the LLM raw execution power in your local environment."

I think this is underappreciated. CLI access gives agents a ton of freedom and might be more effective in many applications. But if you require really fine granularity on permissions -- e.g., do lookups in this db and nothing else -- MCP is a natural fit.