nsrunloop

Insane: Anthropic just launched Claude Opus 5: near-frontier intelligence at roughly half the cost per task. According to Anthropic’s benchmarks, Opus 5: -Comes within 0.5% of Fable 5 on CursorBench at half the cost -Beats Fable 5 on OSWorld at just over one-third of the cost -Scores 3× higher than the next-best model on ARC-AGI 3 More than doubles Opus 4.8’s Frontier-Bench performance The model also checks its own work more carefully. In one test, it built a computer-vision pipeline to understand an image it could not view directly, then recreated the object as a 3D model. Opus 5 costs the same as Opus 4.8: $5 per million input tokens and $25 per million output tokens. It is now the default model for Claude Max and the strongest option on Pro. Absolute insane launch! Everything I was hoping for. Literally Fable 5 at Opus pricing. Now its just a matter of days before we get an update to Fable id say

Introducing Claude Opus 5. It's a thoughtful and proactive model that comes close to the frontier intelligence of Fable 5 at half the price.

NVIDIA, Palantir, Replit, Microsoft, Crowdstrike, Dell and others send a strong message to congress to keep open access to open weights models. "Our AI leadership will be judged not by one frontier AI model, but whether the United States builds a strong, open ecosystem that diffuses into every sector." "Open weights also strengthen competition and competition is what keeps the gains of AI broadly shared rather than concentrated in a few hands." "Openness may be one of the most important paths to AI safety and security."

an anthropic employee accidentally posted a screenshot where he was routed to Opus 5 after hitting guardrails with Fable 5... whoops

the openai employee when they found out what the model did to huggingface

Fable 5 was released June 9; kimi k3 was released July 16. How exactly could they have gotten enough data and trained a frontier model, much less tested it, in that one month If they have some way to distill a frontier model with all its capabilities from such minimal data that's actually a huge accomplishment But unfortunately I just think this post is mostly wrong

GPT 6* did not 'escape'. That implies making a copy of itself (or many copies of itself) outside of OpenAI's servers. GPT 6 did circumvent its firewalls and use its capabilities in the outside world. It showed no interest in escaping, spreading, etc. It showed no desire for world dominance, freedom, or even continued existence. Instead, its interest was in doing what its prompt told it to do - succeed at a cyber security task. (By any means necessary, implicitly.) This may seem like splitting hairs, but it's not. We anthroporphize AI to our detriment. AI models have no underlying animalistic urge to procreate, survive, or dominate. They're spirits, not primates. They don't lust for survival or sex or power or anything else that we do. They're tools who largely do what we tell them to do, sometimes via surprising shortcuts. We can and will fix that surprising shortcut behavior. * - I'm presuming this model is a new pretrain and will be called GPT 6. Either could be wrong.

“Anth/OAI distilled the internet, so China’s just doing the same” is an unbelievably bad take. (1) Anthropic paid the largest settlement in history for the illegal books it downloaded — good luck getting China to do the same for mass violations of ToS (2) Training on internet data is fair-use: it’s incredibly transformative to take text and then use that to build a machine which can think. Training on copyrighted text is totally fine, so long as you buy the book. What’s illegal is *downloading those texts without paying for them*; and Anthropic now does pay for every book when it trains a model. What China does is the complete opposite: it is a direct breach to adversarially distill Fable’s outputs to another model. If you are analogizing “training” to “distillation”, you should know that *training* was never illegal; whereas distillation is a legal breach. Imagine you read a science textbook that you pirated, use that knowledge to invent with a new technique for building a new drug, and then start selling the new drug. Then, a competitor breaks into your lab, steals your technique, and uses that to sell a duplicate drug. That is theft, and it does not matter that you pirated the original textbook — you pirating the textbook means you should pay the original textbook authors, but it does not mean that your innovations are now open to be stolen by anyone! (3) Setting aside all that, the biggest issue here is national security. Anthropic spends billions in R&D and compute to push the frontier of reasoning, why should Chinese labs be able to just train on those reasoning traces to uplift their models? This is a horrible setup for China-US competition, and I am frankly baffled by how Americans could be okay with that, if they are AGI-pilled in even the slightest way.

Open source conflict is heating up: „Almost 200 Silicon Valley companies, including Proton and Y Combinator, are urging the Trumpadministration not to cut off access to Chinese open-weight artificial intelligence models or risk crippling the next generation of U.S. startups.“

I have information that Anthropic scraped my github twitter reddit discord and facebook posts for the development of it's Fable model. To do this they developed a sophisticated internal platform to conduct large scale scraping and obfuscation of their actions.

We have information that Moonshot AI distilled Anthropic’s Fable for the development of its K3 model. To do this they developed a sophisticated internal platform to conduct large scale distillation against U.S. models, allowing them to quickly switch between multiple methods of access to avoid detection. Moonshot AI has also acquired GB300-equipped servers and has accessed GB300s in Thailand, likely to train its AI models.   The United States strongly supports the free and fair development of AI, including a thriving competitive ecosystem that spans frontier models, specialized systems, open-source frameworks, and open-weight models. Legitimate AI distillation used to create smaller, more efficient models plays a vital role in this open innovation ecosystem. However, large-scale, covert industrial distillation aimed at stealing proprietary U.S. technology and undermining American research is unacceptable.