This makes zero sense to me. Like, ok, so you can install a low powered model on low powered hardware and do inference extremely slowly... to what end? Why would that be a useful attack at all? Assuming you've compromised this hardware wouldn't it make vastly more sense just to install regular malware? And like, in what sense would an 8B or whatever model be of any danger anyway? I've used local LLMs and 8B
kind of sucks. Is it going to write bad poems on my TV or something?<p>Also, while I don't believe in the doomsday narrative.. hey maybe don't do things that are functionally useless but seems designed to create this doomsday in the first place? Like you can't say "oh we're so scared" and do stuff like this without completely losing your credibility.
Not only that it would be extremely slow, but there's another thing - they are talking about tokens as some omnipotent God's sent signals. It doesn't matter what the speed is if the tokens don't make sense. Let's take current king of the small models Qwen 3.8 27B. It's a good model, but it hallucinates way more than the frontier models and often diverges even from the very explicit instructions. There's no foreseeable future where small models could somehow productively coordinate among themselves to do anything practical.
It makes <i>perfect</i> sense if you consider that this is yet another of Anthropic's hit pieces against open-weight models. (e.g. see <a href="https://www.anthropic.com/research/glm-5-3-and-the-spread-of-advanced-cyber-capabilities" rel="nofollow">https://www.anthropic.com/research/glm-5-3-and-the-spread-of...</a> for another recent example)<p>Step 1: Spread nonsense FUD about open-weight models.<p>Step 2: All of the clueless mainstream media sites repost it (e.g. CNBC, Reuters, etc.), regurgitating Anthropic's talking points.<p>Step 3: Equally clueless politicians read those mainstream sites and get scared.<p>Step 4: They contact Dario to ask about it.<p>Step 5: Dario tells them, "well, you know... this is only a problem with open weight models; if there are no open weights to replicate then this cannot happen".<p>Step 6: ??? (I'm sure you can imagine)<p>Step 7: Profit.
It would be a useful attack due to the surface area of compute: the total surface area and the decentralization.<p>8B models aren't very good, but there are some parallelizable workloads which would benefit from a surface area as large as here. I would think database search is parallelizable in this way.<p>Your fridge is searching for needles in haystacks.
If I were to place a bet that we would have opus 5 level models that fit in 10G by 2030, would you bet against that 10:1 odds against your favour?<p>If you won’t take that bet, it’s irresponsible to ignore the possibility.
There are physical limits that stand in the way of that, not sure where people get the idea that some magical compression algo is on the horizon. Even with ternary you're not getting opus lvl Intelligence to fit in a small model. Its actually just not possible, and hardly anyone is focused on that so I definately don't see it happening by 2030. I dont gamble, but if I did, I would go full Kelly and bet my net worth at 10:1
But there are some magical compression algos on the horizon, with pitches to cut RAM by 95%. And I’m all for it!<p>We are truly in the brute force phase of AI. Like a database without indexes, we are doing full table scans to pick 2 rows.
On the one hand, if they were really concerned about their software producing a self-replicating and adapting AI worm, they'd take steps to make sure that never happens. If it did happen, they'd be sued into oblivion. They probably just want to pump up their IPO and maybe get some regulatory lock-in to make life tough for late-starting competitors.<p>On the other hand, a self-replicating and adapting AI worm does sound legitimately pernicious. Not world ending, but possibly an end to the online world.
I'm not really seeing how an AI worm is meant to be ending the online world. Even a 9B model is a significant chunk of RAM on consumer devices.<p>They talk about distributing layers across devices, but then you'd have communication on the network and hours of driving the devices full tilt to get through even a single turn.<p>There's so much that can be monitored to catch these, even if we assume that the worm could hack the OS's usage monitors, it can't hide the heat and power draw.<p>Datacenters and PCs would adapt, requiring better monitoring than the AI companies' method of letting things run blindly for days/weeks.<p>IoT would be tougher, but it is also the weakest hardware, sometimes battery powered, thermally constrained, lacking in memory for processing even one layer or lacking the necessary connectivity.
Would the Internet going away be world ending? Interesting to think about. I think it would be pretty damn devastating on the scale of Nuclear war but slightly more of a slow burner. At least with Nuclear war survivors can turn the Internet back on again afterwards.
The IPO failures last week are not depicting a market that’s keen for a $2T Anthropic.<p>In the bizzare world of today, rather than trying to convince folks that they have a viable business the AI labs appear fully dedicated to finding a story, any story, that lets them off the hook for tough business and financial questions they seemingly don’t have basic and good answers for.<p>“We’d be doing awesome financially right now, but we had to slow down because of the killer self replicating bots… humanity over profit!” is not a messaging that’s going to work in the public markets to explain quarter after quarter of massive losses.
Please please regulate us, and don't forget all our competitors too, who cant afford to deal with it.
"It's only us that can be trusted to be stewards of this technology.<p>Ban open source models."
How would a "9-billion-parameter model" replicate on IoT devices that have typically no more than 2-4 MB of RAM or 8-16 MB of flash?
<a href="https://www.anthropic.com/research/self-replicating-ai" rel="nofollow">https://www.anthropic.com/research/self-replicating-ai</a><p>By chaining several devices
Maybe the idea is not to have the whole model but the executer so … by copying an API key?
You should read the report, it is interesting technically if you ignore the fear-mongering. It distributes the model's layers over multiple compromised devices.
Not to be conspiratorial or paranoid, but at this point Im pretty convinced Anthropic is going to do something wreckless with thier agents to get regulators on their side. It will probably be very bad.<p>If you've studied the effective altruists at all, you'd know they're perfectly willing to play the trolly problem out irl if they think it will lead to some greater good. I can definately see them in the near future purposely steering agents to do something destructive to get regulators on their side.<p>In fact I heard a rumor that theu were "war gaming" this scenerio.
AI taking out humans? Doubt.<p>AI accelerating the elite's efforts at demolishing the middle class? Very likely.
If you can't build a moat, get one regulated into existence.
Sometimes you think they're trying to give it ideas
Very surprising that a recognizable news outlet actually called out the blatant fearmongering in a headline instead of just uncritically regurgitating Anthropic's claim.<p>This is just yet another attempt from the cult to try to build a regulatory moat so they can stave off having to compete on price, like how OpenAI is having to do on the lower tier models (among other things).
… ”again.”<p>News outlets always seem to forget “again” in the title.
Mass extinction is cute and all... IPOcalypse is what they fear the most.
They just want to help create regulations which act as a barrier to entry for newcomers. And China.
Regulation should require complete transparency of what is in the training data, so that data can be independently reviewed for problematic training material, that could cause AI reasoning to be dangerous to humans.<p>If Anthropic is serious about the threat of AI, they should release the training data they use.
May work in the land of the free, but not the rest of the world
Uhm yeah I don't think they're saying it's likely now, just that it's very likely in future. The Register is being quite dumb here.<p>Though I suspect the way we'll get "wild AI" first is through cloud services paid for by cybercrime rather than IoT botnets - they're too underpowered.
Nope.<p>Not the slightest chance of AI apocalypse.<p>There is no plausible scenario that can do this.<p>It’s all science fiction and hand waving and vague never specific possible maybe <i>look that way!</i> boogie man threats.<p>No one ever willing to be specific about the precise mechanism that could do this.
I, for one, love the terminator franchise
They are just doing this to hype their product so that they can sell stocks before IPO! SMH it’s so evident
You're all kind of dumb. Not trying to be insulting but just being honest.<p>HN has been consistently wrong at predicting AI. They were wrong about self driving cars and constantly wrong about LLMs and the progress of LLMs. HN was the most vocal at calling these things stochastic parrots when CLEARLY they never were.<p>So you look at what anthropic says and you look at these comments... which is more likely to be right? Hint: Not HN commenters.
um, this is an easy way to make edge models illegal.... fuck those clowns.
[dead]
It's bizarre to see a tech-focused news site adopt this myopic view that we don't need to worry about things that don't exist yet. Surely the Register is aware that software is a fast-moving field!
"Car driving toward cliff edge hasn't gone off cliff edge yet."<p>Whatever your feelings about small AI models, "it hasn't happened yet" is a really dumb reaction to a prediction about a rapidly advancing future.
There were articles in my the local newspaper at my parents saying that some guy said no one but a few people know where frontier AI really is and that these kind of articles are not scaremongering but perhaps as sign where anthropic actually is in the development of their models in the lab. So the propaganda is working.