It kinda seems like a lot of the maximalist plans required the rest of the world standing still. As opposed to other governments, companies, other people generally adapting and trying to get a piece for themselves.
Yeah saw this behavior sometimes. Was fetching some public records from government websites and Claude decided to include my email in the user agent field. Of course, the orchestrator agent said it shouldn't have happened so it's ok.
Some context to this, there's been a bunch of discussion in indie game development spaces about how it might be risky to release demos, or any information really, since a bunch of (probably? can't necessarily assume now.) people seem to immediately try to clone it with agents. It's to the extent I've seen genuine confusion about which game is the original in Twitch streams/chats. There have been a bunch of takes one way or another on it, including 'if your game is clone-able from a demo your idea isn't unique or complex enough'.
True, they don't own the word open. Words mean what people agree they mean, and insisting on more expansive definitions of open got the ecosystem from restrictive licenses in the llama 2 and 3 days to most noteworthy models being MIT, Apache, or OpenMDW (permissive but specifically for model weights).
It's not open weight, it's weights available as they use a modified MIT license that is not OSI. It's the common "if you're serving it as a business you need a license" difference, for what it's worth.
I work at a university, and part of the job is helping researchers with AI. Our bio and chem researchers are starting to despise Claude because it's been actively getting worse over the last months with increasingly aggressive guardrails. Many of them gave up because they feel they can't do anything useful with it anymore. There's a verified life sciences program, yeah. Is it an easy, transparent process capable of handling the hundreds at just our institution? Not even close. Best ad for open models though, thanks Anthropic!
Well, this is certainly different, can't say they ignored the criticism. Wonder if this is a monkey paw curls moment; if the findings continue to accelerate, most of it just isn't public now.
A theory I've been playing around with is that they can't justify to themselves why they're deserving of their immense power and wealth and others aren't. If the real answer is they got lucky by circumstance and aren't "special", that their efforts aren't wholly responsible for their results, they'll twist themselves into pretzels trying to avoid the realization and moral implications.
I usually try not to judge, people have family, commitments, gotta put food on the table. But. I genuinely wonder how people working at some of these companies that actively make the world a worse place live with themselves. This is basically, and obviously, child abuse in the implied usage. Someone might make a tablet good for education, but I don't think Amazon is structurally capable of doing it.
I've been subscribed to Jetbrains full product pack for years now, but nowadays the thing I want more than anything is either a convergence of all their IDEs via plugins or a lighter universal editor like Fleet was getting close to. Agents let me work with languages I had no business with before, but now it isn't unusual to have RustRover, Rider, Webstorm, and Intellij open at the same time. Alternatively, I'm curious as to whether their code intelligence over MCP helps coding agents. If they did, I'd still pay for good code intelligence even if it was a headless MCP server. Thought they were going that way with Air, until they killed it yet again. Air is a plugin now, but that's back to the specific IDE thing.
I hate to devil's advocate this. But is the observed behavior of 'is this an outgoing link' the only practical way to defend against external bot advertising campaigns? Aside from Facebook obviously wanting you to pay them for advertising, how does any independent forum deal with this at scale without banning links?
This time one spent actually money, well cryptocurrency. With all the vending machine benchmarks, I wonder if there's a self sustaining rogue agent in existence yet. Either hosting weights in a VPS or paying for its own API inference. Or whether there will be one that takes advantage of all the bad IoT security. If either occur, if you squint that looks like evolutionary selection pressure. Keep getting reminded of the Cyberpunk Datakrash storyline, where rogue AI make the public internet unusable.
To me it seems the ultimate question is why would we expect humans to have an advantage at anything in the near future. A few years ago that would have sounded insane, but intellectually I don't doubt frontier models are superior to the average person. Even in the arts and humanities, give your preferred frontier model an interface and a way to "see" its work and it'll make the average person's work look like toddler scribbles. Humans still have some advantages, some places more than others, sure. Working with a designer recently I realized how bad, or maybe average, frontier models were at design. But I highly doubt the vast majority of people will be economically viable. Maybe this is one of the great filters, a society driving itself mad when freed from necessity of economic utility and without the big philosophical questions having answers.
They might say that, but it's a bit of a fig leaf next to the implications of everything else they say. They incessantly complain about open weights models, framing them as dangerous and national security threats. The audience for national security rhetoric isn't HN or industry, it's government.
Digging in the model card, "Gemini 4 is the next generation in the Gemini series, a suite of highly-capable, natively multimodal, reasoning models. Gemini 4 Argon delivers frontier performance..." If they're using noble gases there are 2 obvious lighter naming candidates and 3 larger (or 4 if you count Oganesson). I'm not seeing them claiming Argon is the top/biggest/bestest model among the Gemini 4 series, just that it has frontier performance.
Edit: noble gases not noble taxes, though the latter would be a funnier naming scheme.
If you read the terms of service carefully, it seems to suggest they only prohibit you distilling outputs, you should be able to publish them no problem. I'm surprised there isn't a large scale open source project for cataloguing and scoring transcripts. Maybe there is and I just don't know about it. Businesses might not want to publish transcripts of proprietary software, but open source projects? Hobby projects? Seems everyone that's not in the club has a shared interest here.
I have a feeling "intelligence too cheap to meter" is going to haunt/mock this industry. Recall that only a year and a half ago Llama 4 flopped so hard Meta rebooted their development efforts. Desperately wish more information was available publicly about proprietary model architectures and training regimes. Are they running Manhattan project style compartmentalization? Otherwise, are these details, and the economic implications, an open secret in their circles given the amount of crossover between companies? It'd be of great interest to us mere mortals that have to deal with the consequences of how this turns out.
There's a feature in most enterprise email, say Outlook, where you can delegate access to an inbox/address without sharing creds. Very common and normal use case, either for assistants/secretaries, common/shared inboxes, that kind of thing.
Was surprised at how messy and uncurated the release was/is. I frankly expected a major drop of like the Riemann Hypothesis or something from another lab yesterday. Cause otherwise they just dumped a pile of proofs of varying quality and let everyone else figure out whether they're right.
In Stellaris you can play as a civilization of robots who keep their biological creator race alive as "bio trophies". The bio trophies don't do anything meaningful besides by existing satisfy the need their ancestors placed in the robots to take care of them. Starting to wonder if that's the best we can hope for, if these things will be, if they aren't already, better than us at anything that matters.
The common timing is bugging me. The trajectory doesn't seem to have been surprising over the last year, so why these exits now? Hey, anyone on the inside, did y'all secretly figure something out, got a computer god locked in the basement? Are rats fleeing a sinking ship? Please share with the class.
According to Artificial Analysis, one metric is standing out significantly: hallucination rate. Beats frontier models by a good margin at 15%, while latest OpenAI are in the 40s-50s and Anthropic in 60s-70s (mostly). Other near frontiers are closer, Grok 4.7, GLM5.3, and Muse Spark 1.3 are all around 30%. Only other model I recall getting close was Minimax M3 at 18%.
Maybe I'm being pedantic, but GLM 5.3 Flash is not a smaller version of GLM 5.3 as claimed. Despite the name, they're entirely different archs and pretrains. 5.3 is a further post train of 5.2, 5.3 Flash is multimodal from an entirely new pretrain lineage.
Genuine question for people in the space, are the no training and zero data retention agreements likely legitimate or not worth the pixels they're displayed on? I'd assumed the first party ones were worthless, but are the ones for serving proprietary models via AWS, GCP, or Azure more reputable? I don't know the shape of deployments, kinds of access, etc. so I'm curious.
Pet peeve on Google's AI rollouts: there's no alignment across the three platforms they have, consumer, prosumer, cloud. Scroll to the end of every release, including this one, and you'll see different availabilities. The fun part is the models don't even have the same capabilities across platforms! Omni Flash, last I tried and read the docs, is video and text out on consumer and prosumer but video out only on GCP. So if your org disables consumer and prosumer, like mine, it's a coin flip whether you can use the fancy new models or what they can do.
I've followed a few trackers, eg https://marginlab.ai/trackers/claude-code/ , for awhile. For Claude Code the trend, it seems to me at least, is fewer tokens to do the same or better job. Prompt changes, tool ergonomics changes, etc.; I'd be shocked if they didn't A/B every release. Less thinking as measured by tokens isn't necessarily bad if you can get the same results by making it think about the "right" things or structure. They obviously screw up sometimes, and I've always been suspicious with hidden tokens, but I haven't found evidence quality intentionally degrades over time.
"It is difficult to get a man to understand something, when his salary depends upon his not understanding it."
"If this view takes hold, it will shake the foundations of our society"
To me the biggest gap in credibility is the criticism of circular reasoning while his argument is identical but flipped on burden of proof and cost of being wrong. I struggle to entertain the categorical claims, that are very convenient for the status quo and those who benefit from it, with the, at the moment at least, unknowability of anyone or anything else's subjective experience.