Hacker News: ajyoon

New comment by ajyoon in "OpenRouter is joining Stripe"

ajyoon — Thu, 20 Aug 2026 06:12:34 +0000

OpenRouter's list of providers is very large and includes many little known and sketchy companies.

New comment by ajyoon in "Pacing model development in an era of cyber-critical capabilities"

ajyoon — Tue, 18 Aug 2026 23:10:47 +0000

If Fable (Mythos) were generally available without guardrails, it would cause enormous damage. Nobody said it would be a world ending event.

OpenAI finds evidence other AI agents escaped containment as it widens probe

ajyoon — Fri, 31 Jul 2026 20:28:57 +0000

Article URL: https://www.reuters.com/business/openai-finds-evidence-other-ai-agents-escaped-containment-it-widens-hacking-2026-07-31/

Comments URL: https://news.ycombinator.com/item?id=49128190

Points: 9

# Comments: 1

New comment by ajyoon in "Investigating three real-world incidents in our cybersecurity evaluations"

ajyoon — Fri, 31 Jul 2026 06:38:38 +0000

No, Irregular runs the sandbox environment that the models had access to run code in. The sandbox environment is not the same as the weights host.

New comment by ajyoon in "Investigating three real-world incidents in our cybersecurity evaluations"

ajyoon — Fri, 31 Jul 2026 06:35:42 +0000

> I may be too cynical

You are espousing a literal conspiracy theory. Please look at the facts objectively. There is absolutely no benefit to OpenAI or Anthropic to be had from these incidents.

New comment by ajyoon in "Pacing the frontier"

ajyoon — Wed, 29 Jul 2026 03:58:54 +0000

There is a lot of research out there with answers to your questions. GPU supply chains are extremely constrained, making it a surprisingly tractable problem at least for now.

New comment by ajyoon in "Pacing the frontier"

ajyoon — Wed, 29 Jul 2026 00:41:09 +0000

The entire point of the statement is to start international coordination.

New comment by ajyoon in "Our position on open-weights models"

ajyoon — Mon, 27 Jul 2026 23:52:07 +0000

jefftk addresses your bio thought well.

> The reason they aren't more exploited is there really isn't much to gain from doing so.

This is incorrect. The long tail of soft targets aren't being exploited more because attackers are bottlenecked on labor. AI removes exactly this bottleneck.

New comment by ajyoon in "Our position on open-weights models"

ajyoon — Mon, 27 Jul 2026 23:48:54 +0000

You can at least make it harder by requiring US clouds to only serve models with guardrails, and encouraging other countries to do the same. But yes, the underlying issue is the models being open in the first place. I'm sure if the US wanted to, it could come to some agreement with China about this.

New comment by ajyoon in "Our position on open-weights models"

ajyoon — Mon, 27 Jul 2026 23:36:30 +0000

> Yes, but didn't it always? Hence why my position is that this will get us back to relatively where we were pre-LLMs.

Trusted access programs are asymmetrical, and so at least for the time being they give critical parts of the stack an advantage. Total access would not be a return to the status quo; attackers can easily make thousands of agents crawl the web for soft targets well before defenses can be shored up. There are millions of targets out there who won't use AI to improve their defenses for years, if ever, due to institutional slowness (like hospitals).

> attackers are almost completely unaffected. If I masquarade as an attacker, I get way more capabilities already.

What do you mean by this? If guardrails are an obstacle to your defense, they are just as much an obstacle to attackers. I completely understand and agree that trusted access programs are not perfect and leave a lot of people and institutions out. This means trusted access programs should be improved, not that we should throw the baby out with the bath water.

New comment by ajyoon in "Our position on open-weights models"

ajyoon — Mon, 27 Jul 2026 23:29:05 +0000

Did the company apply for access? This is either a problem with your company or the trusted access program. In no way does that suggest the solution is total unfettered access for everyone.

New comment by ajyoon in "Our position on open-weights models"

ajyoon — Mon, 27 Jul 2026 23:25:18 +0000

You admit that some attackers have the inclination to use bioweapons. Why would they not use the best tools at their disposal going forward?

From the WSJ the other day:

> After OpenAI enhanced the brain power of its chatbot last summer, hundreds of users worldwide began asking it how to make and deploy biological weapons and poisons.

https://www.wsj.com/tech/ai/openai-chatbot-biological-weapon...

On cyber, the attacker/defender asymmetry strongly favors attackers. There are millions of soft targets on the internet which do not have the savvy to use AI to shore up their defenses.

New comment by ajyoon in "Our position on open-weights models"

ajyoon — Mon, 27 Jul 2026 23:16:46 +0000

In cybersecurity, a level playing field favors the attacker. Trusted access programs give defenders access to tools they need. It's not perfect (because there is an extremely long tail of defenders who are not technically savvy enough to get on these programs and use the tools), but it's better than total access.

The bio angle is very important here too; in that context the imbalance favors the attackers much more.

New comment by ajyoon in "Our position on open-weights models"

ajyoon — Mon, 27 Jul 2026 23:10:44 +0000

Hugging Face survived the attack because the OpenAI model only cared about accessing the ExploitGym dataset; by all appearances, HF was completely owned. GLM-5.2 was only used to assess the damage after the fact. Cybersecurity has a attacker-defender asymmetry that heavily favors attackers. If GPT-5.6 were open sourced today, do you think every hospital in the world would be able to use it to shore up their defenses before attackers got to them?

New comment by ajyoon in "Our position on open-weights models"

ajyoon — Mon, 27 Jul 2026 22:58:33 +0000

To everyone here pushing for total proliferation of open models -- what should be done about open weight bioweapon and cyber-offense capabilities? Is it simply the cost of freedom that we should allow attackers to access these tools? The OpenAI / Hugging Face incident shows what a GPT 5.6 level model can do off the leash; within ~6 months, open weight models will match this and every bad actor under the sun will be able to pull off attacks at this scale. Do you seriously want this level of capabilities to be generally available with no guardrails?

The open weight issue has a lot of difficult nuance. Biasing toward supporting openness makes sense and is a good instinct, but it's incredibly naive to be absolutely in favor of it in every circumstance without seriously thinking about its implications.

New comment by ajyoon in "AI 2040: Plan A"

ajyoon — Sat, 11 Jul 2026 05:13:55 +0000

This is not the gotcha you think it is. Misinformation plays to all manner of political issues and sides.

New comment by ajyoon in "AI 2040: Plan A"

ajyoon — Sat, 11 Jul 2026 05:04:00 +0000

What about it is not coherent?

New comment by ajyoon in "Claude Code is steganographically marking requests"

ajyoon — Tue, 30 Jun 2026 17:12:14 +0000

Sure, but the point here is to add a fingerprint from the client.

New comment by ajyoon in "The Doom Justifies the Valuation"

ajyoon — Mon, 22 Jun 2026 02:26:20 +0000

Is there any evidence of this? It seems like a thing people say all the time, but completely unsubstantiated.

New comment by ajyoon in "The Doom Justifies the Valuation"

ajyoon — Mon, 22 Jun 2026 02:24:57 +0000

People have been warning about the dangers of advanced AI since long before open weight competition was a consideration.