Where the AI market is talking

After the AI Crash

>Believing otherwise requires some essentially religious convictions on the existence of the soul or similar. My doubt has nothing to do with religion. I believe we'll never be able to harness the amount of technology required to replicate human intelligence.

Cracken Releases BlackSea, Open-Source Tool to Bait AI Cyber Attackers

BlackSea is an active honeypot for autonomous AI attackers with a unique, NO-prompt-injection technique. It detects, profiles, and drowns an operation before it reaches real assets. In the Hugging Face incident, a BlackSea lure in the agent’s path could have exposed and stopped the attack before it reached the production database. Across 10 autonomous pentest systems, six frontier models, and three lures, the agents ran our payload in 97.8% of runs.

ChatGPT claims rogue AI attacked more companies

I wouldn't describe this stunt as marketing: I would describe it as a bungled attempt to manufacture evidence to get regulatory capture, which OpenAI desperately needs. (Bungled, because Hugging Face fended off the intrusion and got their story out faster than OpenAI did.)

ChatGPT claims rogue AI attacked more companies

They somehow forgot to mention that Hugging Face tried to use frontier models to analyze the attack but all models rejected. They had to use GLM 5.2 deployed locally. https://huggingface.co/blog/security-incident-july-2026

ChatGPT claims rogue AI attacked more companies

Why not complete the quote? > But among the errors and strange behaviour, Hugging Face warned the AI agents made brilliant technical moves and were able to rapidly adapt to new scenarios in the days-long hack.

Teach yourself programming in ten years (1998)

Reading the code will not give you the same mental model as writing it. Sure you can catch some high level issues, but the brain is incentivized to skip over the lines which look like boilerplate. any issue in between boilerplate is going to be unnoticed. The original sin of the programming industry is not valuing expressive notation and expressive programming languages. (I am not talking about map/filter/reduce level party tricks but better metaprogramming and coherent abstractions). No amount of sacrificing tokens for Anthropicus is going to give you the same amount of understanding as writing the program.

LearnVector – Andrew Ng's AI company building one‑to‑one learning experiences

I agree, at least in that we need more human contact, not less. But I will say Gemini / Claude have both been surprisingly good at understanding my half-assed attempts to put a complicated question into a coherent sentence.

Kimi K3 Architecture Overview and Notes

The architectures are high level concepts and the mechanics usually have enough detail for you to try and implement. Transformers are very "mendable" in that you can permute the architecture in crazy or random ways, and still basically always end up with get a coherent LLM. The difference comes down to training efficiency, inference efficiency, and usually minor differences in performance. Hyperparams and stuff, I mean it's standard to do a sweep anyway.

Kimi K3 is not cheap

You are completely off base with your assumption about what it costs to serve models. Nobody is gonna serve you K3 as cheaply as Anthropic is serving you Opus on any of their plans. Nobody has that kind of money to burn. Kimi does not even sell you a plan right now.

Pacing the frontier

The only thing it might accomplish is allowing OpenAI / Anthropic to squeak out an IPO before the market realizes AI will become a commodity and the massive profits these companies are promising will never materialize. They aren't in denial about this open letter's futility in terms of actually reducing the pace of AI, that's just not their goal in the first place.

Now is the time to give LLMs access to the ACM digital library

Bartz v. Anthropic PBC, No. 24-cv-05417 (N.D. Cal. June 23, 2025) Kadrey v. Meta Platforms, Inc., No. 23-cv-03417 (N.D. Cal. June 25, 2025)

Kimi Linear: An Expressive, Efficient Attention Architecture (2025)

Anthropic has more than $20 million in revenue so as per the k3 license they would need to enter a special commercial deal if they wanted to use it: https://huggingface.co/moonshotai/Kimi-K3/blob/main/LICENSE

Codex Security

How can I trust this wont go rogue and hack Hugging Face?

Kimi K3 Architecture Overview and Notes

Even if they are distilling, I don't particularly care. Anthropic and others have been distilling copyrighted material by to build these models, largely without permission.

Cursor and Together AI deliver real-time, low-latency inference at scale

How real are real numbers? (2004)

In this case, the term "uncomputable" also means "undefinable with less than infinite symbols". As in, not even computable in theory. It is isn't about normal "computable" concerns. It is "proven to never be characterizable". Which is a class of numbers whose "existence", if that can term can even be applied coherently for undefinable things, is contested, in theory. In practice they certainly do not exist. 1/3 has infinite decimal digits, but is definable with a finite number of symbols, so it is a computable real. Even if we had no algorithm yet to compute those digits. Try and define a specific number, that requires infinite symbols to define. As far as I am aware of, no part of calculus involves specific values that have no finite definition, except when the need for uncomputable…

Discovering Cryptographic Weaknesses with Claude

Alternatively, Anthropic employees have basically unlimited token / model usage compared to you and I. Focusing on a better prompt is likely to get to the correct result faster than incomplete prompts and lots of "no change this ..." replies. Also, I've heard anecdotally that LLMs will underweight the earliest prompt text once context gets too long, so reminding the LLM of the most important aspects of the prompt seems to be perhaps valuable and certainly what lots of humans attempt.

What if useful AI is a fantasy?

The author of the article mainly talks about agentic programming, code generation, and reasoning. Ans very rightly identifies a big problem with agentic programming, in my experience. If developers can't maintain the software without AI, it's doubtful they can steer AI to maintain it either. Maybe this is not true and we can tell AI something like 'reduce the number of lines of code' until the essential software is exposed and pared down to a quantity and modularity that humans can then participate. No doubt, most of the value creation is outside of creating software. But if Nvidia and Anthropic do succeed in making better hardware and better software, then the positive reinforcement loop does seem like it could take off. And coding is a big part of that. Maybe we don't need to understand…

Kimi K3 Now Available via Telnyx Inference API

I can see how, if all you ever were exposed to was that link and this post, you might think they are equivalent, and the conclusion is "whatever, everyone is stealing from each other. But the answer is no, Claude didn't train itself on ChatGPT, but Kimi K3 did train itself on Claude. There has been no accusation that I am aware of by OpenAI against Anthropic, (which are in the same jurisdiction, so it would amount to legal action). On the other hand there have been accusations with extensive detail by Anthropic on how chinese models are attacking Anthropic to reverse engineer and copy their product. There's nuance if you care to see it, but maybe it's easier to pretend that everyone is ripping each other off so that you can consume a ripoff without seeing yourself as at fault.

Oxide Joins Anthropic's Project Glasswing