HNHacker News
TopNewBestAskShowJobs

Jimmc414

16,221 karma · joined April 9, 2019

x.com/JimMcM4 github.com/jimmc414
submissionscomments
Jimmc414··on AI Kill Switch Act: Official Bill Text by Reps. Lieu and Moran (2026)
I don't think a law is inappropriate, and cutting power to a data center is a fair example. But that isn't what this bill does. It requires companies to be able to stop inference, cut off user access and throttle compute allocation. Those are all things they already do to run their own services. Nothing in it touches the grid, and nothing in it reaches open weight models at all.

My concern is with what the government gets, not with what the companies have to build.

There is a provision requiring companies to be able to suspend an account or a user that the Secretary identifies as a risk, and that includes a risk of violating the company's own terms of service. Once an incident has happened somewhere, DHS can order it. The order takes effect immediately, asking them to reconsider does not pause it, they get five days to answer, and after that you are filing in the DC Circuit. What gets reported along the way is exempt from FOIA and from state open records law.

To be frank I worry about mankind and its lust for power more than I worry about the AI. History has not been kind to people who assumed emergency powers would stay narrow or get handed back.

Jimmc414··on AI Kill Switch Act: Official Bill Text by Reps. Lieu and Moran (2026)
This strikes me as very naive at best. Like a super intelligent AI that was somehow intent on going rogue couldn’t figure out how to circumvent a kill switch?
Jimmc414··on US House Narrowly Passes Controversial National Defense Authorization Act
Some of the more controversial provisions:

Section 219 directs the defense secretary to designate a Pentagon “executive agent” to run a US–Israel Defense Technology Cooperation Initiative covering AI, quantum, cyber-defense, weapons research and supply chains.

SAVE America Act attached. The rule merges the voter ID/proof of citizenship bill with the NDAA before it goes to the Senate.

No Iran war powers vote. The Rules Committee killed the Ryan amendment barring FY27 funds for military action against Iran absent congressional authorization.

Jimmc414··on ICE shared Medicaid data it wasn't supposed to have with Palantir
The government's stated remediation for the Palantir transfer was that the data had been shared over a Microsoft Teams chat and was deleted from the chat. That is not a deletion.

According to CMS's security program documentation, Medicare and Medicaid are a covered entities under HIPAA, subject to 60 day notice to affected individuals, reporting to HHS OCR, and notice to media outlets when a breach affects more than 500 residents of a state.

Not to mention this violated a standing court order.

https://security.cms.gov/learn/cms-breach-response-handbook

Jimmc414··on John C. Dvorak has died
Used to watch him regularly on ZDTV. RIP
Jimmc414··on OpenAI reduces Codex Model Context Size from 372k to 272k
There are numerous Claude settings to disable or modify this behavior
Jimmc414··on Flowers for Algernon
Full text:

https://www.kkoworld.com/kitablar/deniel_kiz_elcernon_ucun_g...

Jimmc414··on Secret Claude tracker surprises users after Anthropic's anti-surveillance stance
>Otherwise please use the original title, unless it is misleading or linkbait; don't editorialize.

https://news.ycombinator.com/newsguidelines.html

Jimmc414··on LAPD Regularly Pulled over Innocent People Plate Readers Flagged Cars as Stolen
> Which one of the on-topic comments provides on your first submission.

I take it you are referring to your comment that the URL mistake was somehow intentional. It was not.

I generally take the approach of pinging the mods as little as possible.

Jimmc414··on SCOTUS slip opinion: Cisco granted immunity in building China's "Golden Shield"
Decision effectively ending Falun Gong practitioners' suit alleging Cisco custom built China's "Golden Shield" to identify, track, detain, and torture them.
Jimmc414··on LAPD Regularly Pulled over Innocent People Plate Readers Flagged Cars as Stolen
There's no on-topic discussion there. It was posted an hour ago with a truncated link, discovered after the edit/delete window had passed. https://news.ycombinator.com/item?id=48900001
Jimmc414··on LAPD Regularly Pulled over Innocent People Plate Readers Flagged Cars as Stolen
It was an accidental truncation of the URL.
Jimmc414··on Ben Bernanke Joins Anthropic Oversight Trust
https://www.bloomberg.com/news/articles/2026-07-09/former-fe...
Jimmc414··on Creatine doesn't just build muscle. It may also help fight cancer
Paper:

https://www.cell.com/iscience/fulltext/S2589-0042(26)00811-4

Jimmc414··on Egg consumption inversely correlated with Alzheimer's
Agreed that seems like a potential conflict that should be scrutinized, but it highlights a real problem in health research. Who else would fund a 40,000 participant 15 year study about egg consumption and Alzheimer’s? How many studies are not conducted because no one is interested in funding them?
Jimmc414··on Ask HN: Is anyone experimenting with different ways of using LLMs for coding?
I think the value right now is to focus less on external orchestration if at all. trust the (current best) model to do it better than anything you bolt on to the harness. focus your energy on providing clearer specs. I think the optimal spec is a disambiguated (through liberal use of the AskUserQuestion tool) 1 intent, 2, input/output contracts 3 constraints and 4 preconditions. focus on that and get out of the models way. I think of it like this, imagine a person who was not as smart as you was trying to tell you how to do a task. would you want more verbosity and step by step instructions or would you want them to just cut to the chase (ie, what are you trying to do, what are the obstacles, I'll let you know if I have questions).

also let the model verify itself. don't give it an objective that is vague, give it clear exit criterias for goals and let it loop until it gets there so much of the orchestration scaffolding seems like massive technical debt

oddly, I do the opposite of a lot of conventional advice when it comes to models. I use no memory, I think there is something similar to context rot when everything is stored. I like creating markdown files as memory that the model can grep if needed. I also havent found a real use for hooks yet, I have tried but they always seem to get in the way. skills on the other hand are very undervalued. they are so much more powerful than many realize. I used to think agents were where the power was. I think its actually skills. agents are really for context preservation. skills are what increase capabilities

I'm not even talking about quantity of items in memory, I mean dilution of intent. I really love a model with a clean slate and only the items it needs. I fear the memory guides the model in areas that might not be what I want with the current prompt

progressive disclosure is a big one. you can make context available but it is only loaded when needed. like lazy loading for prompt engineering. skills are to be used to instruct the model how to do something specific that is not in its training data. like how to access my proprietary system, how to interface with a custom program. you can embed templates in skills, you can embed code that executes in skills and only the output is loaded into context. skills expand capabilities, agents constrain context

(constraining context is a very good thing btw, don't mean to infer that agents are somehow inferior to skills)

Jimmc414··on Monetization Gateway: Charge for any resource behind Cloudflare via x402
I'm in awe at how tone deaf and naive the CEO comes across in this article. It reads like a comically ominous punchline from Gavin Belson.
Jimmc414··on Claude Code is steganographically marking requests
I don't see any ethical connection between adding canary tokens to your output to catch people breaking your accepted ToS through ongoing distillation and stealing your PII off of your machine. How are legitimate users in US or China possibly harmed by Anthropic silently changing the apostrophe in Today's or the date separator from - to /?
Jimmc414··on Hackers Broke into Madison Square Garden
Respectfully, the story you referenced is about how MSG compiled "Facial Recognition Activists.docx," collecting its critics' tweets and comments into an internally accessible file.

https://www.404media.co/madison-square-garden-made-dossier-o...

This story is about how the hackers actually got into Madison Square Garden.

https://www.404media.co/how-hackers-broke-into-madison-squar...

I think there is an argument that there are two different stories here.

Jimmc414··on Hackers Broke into Madison Square Garden
https://archive.is/lVFym
Jimmc414··on How much? The hidden costs of restaurant dishes
“On one end £9 of labour cost for a plate of asparagus seems deeply inefficient and unrealistic”

Every time I’ve had this thought that I could recreate a dish and spend a lot less, I end up paying more for the same dish, with lower quality and I still have to do the dishes. It might not be as unrealistic as it seems if you aggregate the wages of all staff involved and divide it by the number of plates they serve over a night. I’m not a cook, but if someone offered me $9 to prepare them a comparable asparagus dish, it doesn’t sound that lucrative.

Jimmc414··on The psychology behind AI fueled delusions
https://archive.ph/0w0KN
Jimmc414··on [dead]
https://archive.ph/gOhKx
Jimmc414··on Anthropic has hired an economist with interesting views on human survival
https://archive.ph/yYHue
Jimmc414··on Fundamental principles of the universe called into question by two physicists
https://archive.ph/5stCw
Jimmc414··on Untethered Goose Game (SoftBank)
https://archive.ph/1D3Qt
Jimmc414··on Fika Jobs raises $4M to build platform where AI agents interview candidates
Agreed. It touches on the same points raised in this recent post,

“If You are Asking for Human Attention, Demonstrate Human Effort”

https://news.ycombinator.com/item?id=48497609

If this interview isn’t important enough to assign a human to the task, it sends a message about what you should expect as an employee. I’d expect a larger percentage of these AI interviewers will be vetting AI interviewees.

Jimmc414··on Die analysis of the 8087 math coprocessor's fast bit shifter (2020)
That's a different article by the same author, Ken Shirriff

https://www.righto.com/2020/05/die-analysis-of-8087-math-cop...

https://www.righto.com/2026/06/intel-8087-adder-reverse-engi...

Jimmc414··on I don't see any good orchestration system for AI agents
The reason there’s no good orchestration layer above the base model is that the useful part of orchestration moved into the model. what’s left to bolt on top is negative value when token consumption, latency and usability are factored in.

I’ve spent a great deal of time trying to beat plain Claude Code and Codex with a variety of planner/critic/decomposition orchestration setups. Every one produces less or same quality, 2x+ tokens, more latency and a second system to debug. External orchestration seems like I’m just creating technical debt and the actual value is focusing exclusively on providing clearer specifications (ie statements of intent, constraints and input/output contracts) to the model and just letting the leading frontier model do its thing with reasoning and effort maxed out.

Ironically the real orchestration benefit is on the human side refining intent to model instead of orchestrator to agent, invoking AskUserQuestion and other tools for the model to more deeply disambiguate my requests.

Jimmc414··on Aikido Code Audit
“But it appears 1 or more organizations have successfully jail-broken Fable 5”

This is hardly true or it’s true of all frontier models and this was only magnified by Fables capabilities. It’s that you could hand Fable 5 vulnerable code, ask it to fix it, return patch plus test cases proving the fix and exploit relevant detail falls out as a byproduct of legitimate secure code review work.

I challenge anyone to provide a fix for this “exploit” without compromising Fable’s ability to patch unsecure code.

← PreviousPage 2 of 33Next →