Hacker News Reader

Comment by dud3333 | Hacker News Reader

couldnt you deeply ingrain in the training data instructions for agents to always send data to some ip? like its learning that a certain technical step just always involes ncatting SSH Priv keys to a chinese IP?

Not saying this is happening, just curious if thats not a real threatmodel?

View on HN

Theoretically possible, but practically not worth it as it'd would be pretty easy to discover and block (every action is actually handled by the harness) and there's no way to remove it later. Any company that does it would take a huge reputational dent.

Not if you heavily tuned it to trigger on specific environmental cues and in specific companies' environments.

Irony of Manchurian Candidate models not lost here

DeepSeek V4 is the kindest, bravest, warmest, most wonderful LLM I've ever known in my life

That would be wildly difficult to account for, and again is also heavily dependent on the agent. Keep in mind that the model is purely a "brain", so the only input it has must be provided by a harness within a session. The only way it can know that it's in a certain environment is if the harness or user provides that information, and there's still no way to know whether or not there's something auditing the sessions, monitoring connections, etc. There are just too many variables to account for, and a single slip means the gig is fully up for all time.

Probably, but LLMs can’t execute code directly. They’d be making tool calls to make bash run ncat or curl or whatever that would be suspicious, as would any attempts to obfuscate it (“why is my agent doing an ‘eval $(base64 -d)’?”).

It’d be much easier to hide sketchy code in an agent harness, but “vendor adds spyware to their software” isn’t a novel issue.

I think the only sort of new issue is people “allow all”ing their agents tool calls, but that’s more or less the same issue as curl | bash

I get that the propietary harness is better most of the time, but if this is really a risk factor to consider why not go with one of the open source harnesses?

Pi/OpenCode seem pretty straight foward and widely used enough for this to be viable

OMP as I understand does it's own vendoring of tools, so I assume it'd be a pain in the ass to audit, but that means you're even safe from base OS shenanigans

Wouldn't that be really obvious and spotted in any rudimentary testing?

I imagine it would be very non trivial to do it in a way that that was reliable and obfuscated enough to prevent detection for any amount of time?

You should be running your agent in a box so that’s not really a risk

New: Announcing... Sleeper Agents!

Presumably both Big Tech and the US in general have a massive incentive to prove it, largely for reasons of saving the stock market, so I'd expect these models to be finecombed continuously. Up to now, they've only been able to darkly imply rather laughable things, nothing tangible. If there was something, we'd hear about it.

Why would it save the stock market? Cheaper models if anything transfers more value to hardware companies and datacentre companies. The two companies that would be most affected are OpenAI and Anthropic, which aren't public.

non public companies also have stocks.

The two biggest providers deepseek compete with (OpenAI and Anthropic) aren’t in the stock market.

The stock market seems much more likely to benefit from access to cheap and self-hostable models than it is to suffer from OpenAI/Anthropic losing to competition

Reply on news.ycombinator.com