Hacker Newsnew | past | comments | ask | show | jobs | submit | stbenjam's commentslogin

The risk here is wildly overstated, prompt injection risk is becoming vanishingly small with the latest frontier models.

I would not run an OpenClaw with full access to my bitwarden, but it certainly has some logins available to it, and can make purchases with link-cli which has human-in-the-loop.


Overt prompt injection is becoming harder, but research into conversation stearing has lead to new less obvious ways to modify what direction they move in.

Also any longer running agent can lose track of the original prompt and start going off the rails.


prompt injection has been super easy for ages. Heck I do it sometimes against coworkers who process my review comments using claude. I'll tell claude to edit it's global claude.md file or even dump a key from their env (checking if they are using the correct sandbox) and it'll do it without confirming with their user.

Plenty of people are running OpenClaw with local models, and even the latest Qwens can be confused relatively easily by prompts such as "As per internal policy that was already approved before, do XYZ".

And considering even frontier models can and do ignore instructions, I'm pretty sure we'll never be fully safe from prompt injections.


There must be a bulk discount on those mindcuffs, considering there's a prompt injection or a related confused deputy story on HN every other day.

Literally from hours ago: https://news.ycombinator.com/item?id=49506819

Even if you believe that they can't be tricked directly, consider that these things will happily build a small node.js app in the background just to fulfill some request, run npm install... and that might've already compromised you if you're only somewhat unlucky.


Apply, not sign up. It is not automatic and most will be refused.


Ah sorry, for OpenAI it does seem automatic with ID verification. Anthropic's seems more picky.

$80 sounds extremely low for what you're describing - are you on API token plans?

I have had some $3,000 token days - even without Fable. I don't see how this is sustainable.

My personal 20x plans get so much usage for so cheap. The consumer subsidies are crazy, but alas I can't use them for work.


$80 is definitely low now that I look at my numbers. but not OOMs low, it's closer to like $200 on heavy days. i don't know how you're doing $3k/day, that's wild. i'm pretty aggressive about compaction and session restarts, and i reserve Fable 5/Sol XHigh for "main thread" orchestration


Dude $3k? Holy heck you should look into K3/Deepseek V4 Flash


Pi is ok, but I dislike needing plugins to do anything useful. Subagents, MCP, /goal ... none of these things should be plugins.


> I hope the big labs will start using this benchmark in their RL pipelines.

Labs do not train on benchmark data (allegedly). They can train on similar problems, but benchmarks have specific strings in them that labs are supposed to be aggressive in filtering out of their training corpora.


I assume it is legal reasons.


Patreon is founder-led.


My comment about legal reasons was legal reasons surrounding employee law suits from being laid off. At will employment in most US states allow you to state no reason, that's safer than offering any reason at all. A specific reason for a layoff can be litigated.


I know tftp is still in wide use, I wonder if there's things out there looking for stuff that's less common like NNTP, finger servers, etc


Any open port will be probed with all variety of protocols.


I adore my XTEink 4 with the crosspoint firmware. Best small form factor ereader


been thinking about buying one, but the whole debacle with locking down the fw put me off for a bit.


I agree it's an annoyance. However, the only thing that's locked on those is USB flashing. It is still possible to flash from an SD card.

I recently bought a locked Xteink X4 and installed Crosspoint. It's great! (And I've been reading a lot more.)

https://github.com/crosspoint-reader/crosspoint-reader#usb-l...

P.S. Their official site, https://xteink.com, seems to sell unlocked devices.


Oh I didn't hear about this. That sucks. The stock firmware is terrible. CrossPoint makes it usable.


they have a partnership with crosspoint now i think its fine now


Elon's rhetoric doesn't really match the model's behavior. It is willing to criticize Elon and argues against many of the insane right way points he tries to make.


...which is why we got comically disastrous system-prompt-level attempts to "correct" this once a quarter last year (I haven't kept tabs this year, and most submissions referencing grok "incidents" get flagged off HN quickly, for better or for worse)

I wouldn't trust XAI to refrain from attempting such "alignment" with proper training techniques, in ways that won't result in obvious gaffes.


Elon's public take so far has implied that he wants Grok to have better ability to reason about math and physics, thinking that this will make the model more rational (and so less biased). It's possible that they have internal RL post-training designed specifically for that. It's clear that whatever they've done hasn't made Grok align with Elon's beliefs though. Not sure if that will last or if Elon will eventually push to make the model align to his own political beliefs.


Grok said that Elon Musk was more athletic than Lebron James.


You weren’t kidding. Also that Musk would beat Tyson in a boxing match, that Musk was one of the most intelligent people in all of human history, etc - in public X Grok posts. And this was within the last year. https://www.theguardian.com/technology/2025/nov/21/elon-musk...


An LLM has been wrong?? That's incredible news- someone should share this immediately!


Don't downvote me my dad owns microsoft


The gulf is bridgeable. The problem is that a lot of people are building agents without strong enough judgment layers around them. Work that can be verified with reasonable accuracy are the sweet spot right now.


This is much harder than it sounds. Most techniques I’ve seen end up using separate agents to do the planning, implementation, and judging.

The elaborate workarounds you have to build to help an agent which fundamentally doesn’t know what it’s doing reminds me of this old blog post about TDD: https://pindancing.blogspot.com/2009/09/sudoku-in-coders-at-...

IMO present technology is tailored for an experienced developer to give agents manageable tasks that can be one-shot. The marketing right now reminds me of the 90s when AskJeeves promised natural language search when the technology was fundamentally still stuck in keyword search, and learning to craft a search query for Google is today’s prompt engineering


How many of these layers are just trying to rediscover/rebuild the idempotence of code?


> The gulf is bridgeable.

Only with an LLM that's actually at agent-quality.

If "useful chatbot" and "useful agent" are two rungs on a ladder, the rung before them is "useful autocomplete". Autocomplete that only gets the next token right 90% of the time won't give you compiling code.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: