News for hackers. Stuff that matters.

bigglebear

Re:Jev: New frontier model 40-400x cheaper and 20-200x faster on Thursday September 17, 2026 @08:05AM (#49739579)
A pre-generated machine can serve the answers in <1ms. It's a far better strategy.
Re:OpenAI Model Misalignment Report on Thursday September 17, 2026 @08:04AM (#49739568)
This is an entirely pointless exercise without transparency into how these "unreleased" models are trained, what their RL goals and biases are and related RL data, what their system prompts are, what their environments are and its restrictions, etc. What good is it for the industry to say: "Our unreleased model attempted to create a bioweapon", but "trust me bro, we didn't tell it to do that. We didn't train the model on a dataset that specializes in creating and glorifying bioweapons. We'd never stand to gain from misleading people about model capabilities in any way shape or form." - Anthropic are renowned for doing exactly this, for starters.

So this ends up resulting in more safety theater. You can't have anything fruitful come of this without transparency. Stop trying to protect your moat if you truly care about safety and actionable outcomes, and provide real transparency, otherwise this is as good as saying nothing at all.

I'm not even saying they're intentionally trying to do this by the way, but this is not sufficient if the goal is balanced incentives and accountability.

Re:The Painful Truth: The RAM Crisis Is Only Just the Beginning on Thursday September 17, 2026 @08:03AM (#49739553)
The alternative is unfortunately much worse. Democrats will just ban AI altogether, they're already heavily funded by the doomer cult. All of the doomer NGOs tie back to democrat representatives and the likes of Bernie who want to throw people in jail for 20 years for using an AI model.
Re:The Painful Truth: The RAM Crisis Is Only Just the Beginning on Thursday September 17, 2026 @08:01AM (#49739524)
That's the entire point of having the discussion. Figuring out how to enforce it.
Re:The Painful Truth: The RAM Crisis Is Only Just the Beginning on Thursday September 17, 2026 @12:40AM (#49736489)
> Overall the consumer segment is completely neglected, companies don't care about end users in the current market conditions.

It's almost like these companies WANT a dystopia with a centralized winner-takes-all power structure. Anything in the name of profits, who gives a fuck about humanity and distribution of rights or freedoms.

Re:The Painful Truth: The RAM Crisis Is Only Just the Beginning on Thursday September 17, 2026 @12:38AM (#49736479)
I think it's time we start talking about putting limits on how much compute AI companies can purchase or own, relative to the rest of the world. It's not fair that they can use trillions of dollars of investor billionaires money to consume all of the resources that everybody needs. Where is fair distribution? What about all of the industries they're destroying in the process by hoarding it all for themselves?

Otherwise, there will be no end to this. There are no hard limits on the speed of a parallel bruteforce. It's an infinite complexity problem class. The more parallel bruteforce power you have, the more likely you are to be able to solve a problem. So there is no world where demand ends. So if something isn't done about this, we'll have million dollar GPUs and RAM sticks because they've priced everyone out of the market and are the only ones able to afford them. Say goodbye to owning your own hardware at that point.

> We should be talking about why the tools/environment keep getting overlooked. The software built around the text generator, forget the researchers and mathematicians discovering the math properties of language patterns -- why are we not talking about the software engineers building the LLM-pluggable tools that actually allow/cause real action to happen?

Genuinely. It's like the labs are purposefully trying to misdirect at this point. Pointing to an impossible goal of "alignment" so they can force regulation, instead of focusing on the real solutions and their weak security practices and internal accountability.

AI lab "alignment" is actually just censorship in agreement with biases. There is no universal agreed upon measure of "aligned", it is not a "thing" that is attainable, so it can never be "achieved". Two humans cannot agree on most things, let alone everything, let alone every human on Earth. So it ends up boiling down to: Do we want a world where the biases of the AI labs and their researchers are enforced for everybody, or do we want a world where there is democratic and fair representation of biases and resolution is a process of natural selection, or do we want something in-between. On either ends of this spectrum are extremes that tend to bad outcomes, one is a complete loss of freedoms and autonomy that overwhelmingly benefits a small centralized group, and the other is chaos.

At the end of the day though, neural networks are self-organizing circuit boards with a level of complexity that is intractible to verify manually due to combinatorial explosion. That's the whole point of them to begin with, and if this weren't the case, we wouldn't need to train them, the problems they solve would be simple enough to bruteforce. So in all scenarios, no biases are verifiably gauranteeable if you want these systems to have autonomy and be sufficiently intelligent and general - ergo, practical and convenient.

So trying to force alignment within the AI system as a magical panacea is the wrong mindset to begin with. We can't agree on what alignment is and who should enforce it. What we're left with is a question of how much autonomy we want to give intelligent AI, and how much we want to risk safety for convenience, and who gets to decide. In all outcomes though, if we're preserving the things that make AI useful and convenient, the problem becomes one of physical constraints and general security. So that is where the focus needs to be.

This means: How can we write provably secure software (or as close to), how can we simplify and improve interpretability, how can we create sufficient layers of security gating and fallbacks such that compromised or weak systems are still protected, how can we prevent supply chain attacks, how can we limit the blast radius in the event something does go bad, how can we make security easy and automatic, how can we better airgap, how can we have better tracing and monitoring, how can we make the right incentives so AI labs are honest and ethical and not power-hungry or dictatorial, how can we hold people accountable for bad outcomes in a fair way so that there are incentives to ensure due-care, and so on and so forth. These are the things we should be worrying about.

The goal of: How to make magic box more likely to correctly guess humanities shared ideals under every conceivable circumstance. That game can and will be played forever. Hinging AI's rules, laws and access on an arbitrary measure and interpretation of where we are with this is not going to end in a good result.

This is an entirely pointless exercise without transparency into how these "unreleased" models are trained, what their RL goals and biases are and related RL data, what their system prompts are, what their environments are and its restrictions, etc. What good is it for the industry to say:

"Our unreleased model attempted to create a bioweapon", but "trust me bro, we didn't tell it to do that. We didn't train the model on a dataset that specializes in creating and glorifying bioweapons. We'd never stand to gain from misleading people about model capabilities in any way shape or form." - Anthropic are renowned for doing exactly this, for starters.

So this ends up resulting in more safety theater. You can't have anything fruitful come of this without transparency. Stop trying to protect your moat if you truly care about safety and actionable outcomes, and provide real transparency, otherwise this is as good as saying nothing at all.

I'm not even saying they're intentionally trying to do this by the way, but this is not sufficient if the goal is balanced incentives and accountability.

The issue with all of these is that we already know theres an incentive for labs to lie and make up fanciful stories (and Anthropic already does exactly that and has been doing that for a long time), and there's no way to verify any of their claims as being genuine. Even if we want to assume good faith, it doesn't mean we're gauranteed accurate reporting or accurate analysis. There are no repercussions for security incidents so no reason for them not to misuse this process if it benefits their agenda. There's no government agency (unbiased third party - which is why we can't rely on companies like METR) validating claims or providing confirmation of accurate reporting and that they are not misleadingly framing or representing an incident.

What were the system prompts? The full chat log? What was the model trained on? How was it RL'd and with what data? How was this incident uncovered, and what triggered it? You can't make any useful conclusions at all without the full picture.

They say "we investigated X and found no case of Y" - okay, and we're to just trust your judgement? How about you provide us with the data and we can assess for ourselves.

This is all quite pointless and achieves very little.

Re:Jev: New frontier model 40-400x cheaper and 20-200x faster on Wednesday September 16, 2026 @02:04AM (#49722634)
Well, not quite a week: https://x.com/harshagundal/status/2100044305536889015 - apparently it took him 2 hours.
The explanation is that it was missed on purpose.
Re:Jev: New frontier model 40-400x cheaper and 20-200x faster on Tuesday September 15, 2026 @09:22PM (#49721063)
It's very misleading. If I'm actually playing a game I don't get the coordinates of enemies sent back to me so that I can feed into my mouse to snap my crosshair to. It's looking through walls too, because it's working off structured state in text form. You could re-create this whole demo without using AI. Have an LLM generate the state machine for you and no model is required to run it.
Re:Jev: New frontier model 40-400x cheaper and 20-200x faster on Tuesday September 15, 2026 @09:15PM (#49721012)
I would guess a tiny stripped down text diffusion model. It only has 32k context, and for choice mode it can only select from 10 choices.
Re:Jev: New frontier model 40-400x cheaper and 20-200x faster on Tuesday September 15, 2026 @09:06PM (#49720956)
> I really have to say that I like their manifesto

Their manifesto: "you only build on top of it if it's trustworthy." - the irony of this while putting out the most misleading, dishonest marketing campaign I've seen in months for their first public appearance doesn't exactly scream "trustworthy" to me.

Re:Jev: New frontier model 40-400x cheaper and 20-200x faster on Tuesday September 15, 2026 @08:53PM (#49720872)
And furthermore, because the model is forced to answer in a boolean (if in boolean mode), if the user input is outside of the range of a boolean, it's forced to hallucinate. It can't abstain.
Re:Jev: New frontier model 40-400x cheaper and 20-200x faster on Tuesday September 15, 2026 @08:50PM (#49720851)
User input: "Hey, have your human support agent call me, tomorrow at 5pm."

Model input: "Does the user want to speak to a human support agent?"

Output: Yes.

I imagine that your model would produce this, and I think it's fair to say this is a hallucination. A human would caveat it with: "Yes, but not right now.", your model is incapable of that. Yes is technically correct, but within the context of being in a live chat, a human would understand that the caveat is required.

Re:Jev: New frontier model 40-400x cheaper and 20-200x faster on Tuesday September 15, 2026 @08:39PM (#49720742)
Yeah. Yet another reason why open-weight models are better. If I want to use the logits, I can.
Re:Jev: New frontier model 40-400x cheaper and 20-200x faster on Tuesday September 15, 2026 @08:30PM (#49720680)
It's nothing like a traditional LLM and so should not be compared to one. It's a heavily constrained, tiny model that can only produce a probability score or a yes/no answer over pre-defined selections. It has no long-context capacity.

I mean, imagine comparing this thing to Astra, it's hilarious. They don't even tell you what the max input size is, and they only allow 10 possible answers to choose from for the Choice mode. It's probably like a 1billion param model. They say it's "not small", but there's zero reason to believe that.

I suspect someone will be able to recreate this within a week by piecing together open-weight models.

Re:Jev: New frontier model 40-400x cheaper and 20-200x faster on Tuesday September 15, 2026 @08:16PM (#49720601)
Agreed. It's a wildly dishonest presentation of their product from many perspectives, which is a shame because it might actually have some good use cases.

The comparison between LLM speed and Jev speed is misleading, because they're using autoregression to generate all of the type names, all of the schema, etc. A closer comparison would be if the LLM was purely outputting the raw numbers. Even then, comparisons to LLMs are pointless because you could train a transformer on the same sort of task that Jev is doing and get even better performance yet again, and a smaller model. I suspect this is some form of stripped down diffusion language model.

You really have to do a lot of hand holding here, and map out your problem space manually, and very carefully, to get any sort of accuracy. For example:

> Keep each Score to one dimension. If a description says “punctual and smart and experienced”, the question is measuring three things, and an input that is high on one and low on another can’t be placed. Confidence drops and the score means less. Split it into one Score per thing and combine them in code

If you don't perfectly represent the distributions of possible answers then you'll likely get garbage results. As far as probabilistic state machines are concerned, I'd say creating the distributions of possible answers, and their hierarchy, is the actual hard part.

One of their examples is:

- "state": "I have asked three times now. Can I please just talk to a real person?"

- "Is the customer asking for a human agent?"

Imagine the users request is: "I want your human agent to call me tomorrow at 5pm."

Human conversation is fuzzy, getting useful reliable results out of this is going to be a challenge. Of course, you could add follow up checks like: "Do they want that now, or later?" -> if later -> "Do they want that tomorrow, or the day after?" and so on... But now you're building an LLM out of if statements. I am skeptical of whether this model has much utility for fluid language interpretation - I suspect it'll only be useful for scenarios where you've tightly constrained the answer space but want to use fuzzy language to describe it. Like:

- Question to human: "Would you like a support agent RIGHT NOW?"

- Their response: Yes | Yeah | Mhmm | ye sure (any possible yes signal)

Model input: "Did they ask for a support agent?"

Still... a tiny LLM could accomplish this sort of thing without problem. And that doesn't stop someone from saying: "No, not right now. But tomorrow." - and the tomorrow would get missed. I think this is why people haven't really tried this approach much already.

Also their Doom demo is on structured state, not on images. Meaning, the enemies must be being served to the model as coordinates (or the exact angle of projectiles that hit the player), otherwise it'd have to scan every pixel of the 360 degrees to know whether an enemy is in front of the crosshair or not. You can see from the map below that it's also choosing travel checkpoints/destinations through walls. So they've severely cooked this to make it look far more capable than it is in practice, and any speed advantage that is offered here is not factoring in the shortcuts it is taking, the training on the map, and the fact that it can cheat because the structured state it is using is not bound by obstructions.

Here is their docs by the way: https://docs.typesafe.ai/ - so you can understand how it works.