Where the AI market is talking

Mistral AI wants to build 1 gigawatt of European compute by 2030

Launch HN: Bullet (YC S26) – A Faster Coding Agent

(Not the author but) To your last question, I was able to replicate some of the benefits of my custom harness in Claude by just adding a custom MCP, startup hooks etc. The issue I ran into is that I was fighting Claude's system prompt, which told it to do things which negated most of the benefits of my setup... For context my startup hook was "grep def", lol. (Now working in js I have grep function, grep class, interface etc.). Gives it a little outline for context, saves time poking around blindly. My repos are small though, this is still a big blob if you have very large repos.

Blog about things you don't understand yet

You raise an interesting point. I agree that discussing a subject with a knowledgeable mentor, AI or otherwise, is a great way to learn. But I think back-and-forth discussions are limited as a means of testing your mastery of a subject. Especially with complex and nuanced subjects, being able to organize them in your mind is an essential part of mastery. You will not know if you have mastered them unless you can generate a clear and coherent description of them entirely on your own. With a mentor prompting you, step by step, you are relying in part on the mentor’s understanding of the subject to guide your thinking. It is like the difference between multiple-choice and essay exams. In the former, the questions you are asked reveal much about the structure of the subject to be explored and…

How AI text watermarking works

Wow that is crazy. Thats very plausible and is actually a coherent thought. I didn’t see that at all. Has claude always been this bad? I feel like its getting pretty fucked lately. Coding is still good.

Gemini 3.7 Flash

I tried to use Gemini for one of my projects a month ago. Immediately after signing up and paying for credits, I got an email saying, “Action required: your billing account {redacted} is past due or has invalid payment information.” I have no idea why it says this. My credit card on file works. My balance updated with a new amount from that card. 11 days later, my account was terminated. I still don’t understand what went wrong or how to fix it. I do pay for OpenAI, Anthropic, and ElevenLabs keys.

Nine PBS sues Iron Mountain over blocked access to archival data

By my calculation using what I can find of 2019 pricing, a more reasonable solution would have been: Synology RackStation RS1219+ - $1,200 8x Ironwolf Pro 12TB drives - 8x $450 = $3600 so for a usable RAID-6 capacity of 65TB about $4800 + OSS, or double that if they have two locations they could and replicate between them. I know there would be cheaper hardware options available (I'm partial to older surplus enterprise hardware myself) but it isn't particularly reasonable to expect a small business to choose a more DIY (at least in terms of software configuration) option. You could step down to 10TB drives, but it seems unlikely they would have wanted to risk that not being quite enough. On top of that, since 2019 is 7 years ago, it is reasonable to expect they would have had a hardware r…

Nine PBS sues Iron Mountain over blocked access to archival data

All this over 50 TB of data. Could have created an in-house replicated NAS with like $2500 worth of hardware.

Codex in ChatGPT desktop app for Linux is now in preview

Codex CLI has support for keybindings tho last time I checked they were quite limited and you cannot replicate full universal text editing experience.

Accelerating GPT-5.6 Sol Ultrafast

In light of recent news, it is hard not to think about the 4.5-day hack on Hugging Face's systems happening ~10 times faster and be slightly concerned.

Mistral AI Strategy

Mistral AI Strategy

Mistral AI wants to turn European AI sovereignty from a talking point into a product — one with a service-level agreement attached.

Grok 4.6

Counterpoint: Our engineers are currently trying to replicate "intelligence" and we think it's like the brain because today we think the brain is where "intelligence" reside, and today we strongly believe that "we" are our brains. I would argue that the inventors of those past technologies were trying to replicate other things (movement, energy, pressure, pneuma, psyche, physis, whatever) that felt deeply human to them, and that in the zeitgeist of his time, Hero of Alexandria (1 BC) could also have said: > I suppose you could argue that none of the stories of the past had the actual philosophers behind them attempting to replicate "pneuma" though. Paracelse was trying to create a homonculus by using semen, manure, and blood in the 16th century or so. A few decades or centuries from no…

Gemini 3.7 Flash

Similar experience here, for what it's worth, though I didn't get as far as you. I basically just stopped and didn't bother - it was easier to go through OpenRouter than spend more energy on it. Also: > Google assumes that they're serving companies at Google scale first So much this. I'm currently grandfathered in until the end of the year on Google's Search API, but the $35,000 they want to continue usage of my < 1000 personal searches per month, not going to happen. It has honestly been easier to use Anthropic to help me build my own search index & crawling infrastructure than deal with Google.

Ask HN: How much money do you spend monthly on subscriptions for AI models?

run a 2-person custom dev agency and we just shipped our own saas, so ai is basically our co-founder at this point. i currently maintain active subscriptions for claude pro and google ai pro. honestly, the roi is completely undeniable when you factor in how much time it saves us on boilerplate code, debugging, and drafting initial architectures. we also sprinkle in some midjourney and elevenlabs for asset generation when we need it. for a small team trying to ship fast, the $40-$60/mo we spend on core llms is by far the cheapest overhead we have.

How Organizations Use AI: Evidence from ChatGPT [pdf]

Would it kill these OpenAI guys to learn how to write a white paper? Google's DeepMind did better with their AGI to ASI, though I have qualms on their execution. Whenever I see these Frontier Labs "publish a paper" on ArXiv.org the phrase Esse Quam Videri strides past my mind—is this genuine progress or merely a marketing metric for their stakeholders?

Ask HN: How can you tell a site/app is vibe coded?

For B2B SaaS apps, you'll notice that the panels and navs (top navigation, bottom navigation, left-navigation, etc) are much smaller (which is a slightly more 'modern' look) - e.g., OpenAI/Claude dashboards. In general components are more uniform/standard, fonts are smaller. Again, this is just what I've noticed for specifically B2B SaaS apps. If you ask Claude Design to build you one - you'll see exactly what I'm referring to. I honestly don't mind it. For these B2B SaaS apps its nice to have some site consistency. It almost reminds me of a a new "Bootstrap" way of doing things.

Gemini 3.7 Flash

I didn't have that issue back when I originally created my API keys a couple years ago. Just out of curiosity I switched to a different Google account that had never interacted with AI Studio and never used Google Cloud console. It was literally two clicks, and didn't even leave the page: the dialog asked to create a project and type in a name, I did that, clicked submit and then it was selected as the default project. One more click and I had the free tier API key. Not saying you didn't have that experience at the time, but personally I have had zero issues with AI Studio and consider it the most dead simple/fastest dev dashboard to get started compared to the others like OpenAI/Anthropic (thanks to Google's free tier that lets you skip billing setup annoyances just to play aro…

Gemini 3.7 Flash

I didn't have that issue back when I originally created my API keys a couple years ago. Just out of curiosity I switched to a different Google account that had never interacted with AI Studio and never used Google Cloud console. It was literally two clicks, and didn't even leave the page: the dialog asked to create a project and type in a name, I did that, clicked submit and then it was selected as the default project. One more click and I had the free tier API key. Not saying you didn't have that experience at the time, but personally I have had zero issues with AI Studio and consider it the most dead simple/fastest dev dashboard to get started compared to the others like OpenAI/Anthropic (thanks to Google's free tier that lets you skip billing setup annoyances just to play aro…

Donkey.bas is 45 Years Old – 131 line of Glory

Clearly this needs to be replicated as GTA minigame.

Gemini 3.7 Flash

You prepay for tokens exactly like the OpenAI/Anthropic dev dashboards when using AI Studio, which the link above is pointing to not GCP, also there are project specific spend caps now. https://ai.google.dev/gemini-api/docs/billing#spend-caps