DerangedBytes

A semi-reformed deranged byte.
Hands-on local AI: quants, setups. AI agents in real business ops. Occasional security takes. No hype, No fluff, just numbers

[DerangedBytes @DerangedBytes]

Excuse me, Claude did what ?! 🫪

[Gabriel Ocana-Santero @OcanaSantero]

Last night an AI agent performed brain surgery on a mouse.

We gave it 19 gene therapy vectors in a spreadsheet, a mouse model of epilepsy, access to a motorize surgery rig with cameras and one goal: reduce seizures.

It picked the drug. It picked the brain region. It derived the coordinates from three papers. Then it drove the stereotaxic arms into the brain and injected 250 nanolitres while watching the entry site through the rig camera for bleeding.

A licensed vet gated every irreversible step and held a physical kill switch to stop the AI agent.

Thread on what it decided, what it saw, and the moment it refused to let me override it.

[Writing this on the spot out of excitement, but I've been 18hrs in front of my laptop coding and coordinating this, so keeping it brief and descriptive, more cohesive thoughts on this soon...]

[Tibo @thsottiaux]

Day 25/

We are doubling all usage limits for all Pro 100/200/500 tiers and Business Premium, effective today.

We heard your kind feedback and won’t be reducing usage limits for Pro 200, so your Work and Codex usage will stay the same. We’re also going to continue offering extended limits for dots for now, and usage of your dot will still continue to be separate from your default usage.

And we'll of course also keep shipping better models and other efficiency improvements that increase your effective usage over time.

[DerangedBytes @DerangedBytes]

@RogerKernel O grok, mismo resultado

[DerangedBytes @DerangedBytes]

@RogerKernel Grokbot

[DerangedBytes @DerangedBytes]

@chiefkittenme @bot 🫪

[DerangedBytes @DerangedBytes]

AI gives mega ai companies leverage, not us. And that’s the problem. Regulation is not about protecting jobs or whatever mr.sanders cries about it’s about making sure those companies don’t turn into demi-gods and eventually enslave anyone who depends on them which is more than a feasible scenario to happen than ‘universal income’ because this is planet earth not narnia

[DerangedBytes @DerangedBytes]

@richlonbay Hi there rich 👋🏻

[DerangedBytes @DerangedBytes]

@theo All of the above. Plugins?

[Theo - t3.gg @theo]

It is insane to me that there are still devs using terminals to prompt their agents

[DerangedBytes @DerangedBytes]

Never leaving this app 🤣

[OpenAI Developers @OpenAIDevs]

Starting today, eligible ChatGPT Pro and Business subscribers can claim up to $500 per month in API credits to build apps and agents.

[DerangedBytes @DerangedBytes]

Interesting post from the Chinese community

[make claude @make_claude]

中国社区逆向拆解了Claude的封号机制。以下是他们的发现

Claude使用积分制。注册时账号获得初始分数，之后根据行为加分或扣分。低于阈值直接封号。回分很难

影响分数的因素:

注册
> 干净IP是最重要的因素。不一定要家宽，但必须干净
> Gmail已经不是最优选，有权重更高的邮箱服务
> 注册时直接弹KYC的号可以扔了
> 用手机App注册会加分

支付
> 刚注册就订阅会扣分
> 直接上Max会大幅扣分。最好等两天先订Pro
> 支付时弹KYC说明账号在考察期，但不一定会封
> 国外实体银行卡加分。Google Play也行。虚拟卡扣分

使用
> 不要一次把限额用完。宁可两个号轮换也不要一个号撸到底
> Desktop客户端有少量加分。尽量少用网页端
> 手机端也有少量加分
> 开启记忆功能加分。记忆文件越多分越高

IP和设备
> 稳定性是关键。不要跳IP，尽量一直用同一个
> 代理必须全局模式并开启TUN
> 多设备登录大幅扣分。理想状态是一台电脑一个客户端
> 频繁换IP、频繁登录、高并发都会大幅扣分

账号年龄
> 老号有巨大的容错空间。2023年的号明显更高
> 分数随时间增长，但很慢

其他
> Windows和macOS没区别。Linux会扣分
> 时区和语言基本不影响。不会因为北京时间就封号
> 跟Claude说自己是中国人没问题，记忆不会触发封号
> 大量使用Fable会扣分
> 升级Max大概率进入观察期，除非是高权重老号

观察期和封号
> 接近阈值时账号可能进入观察期。典型特征是掉登录
> 观察期继续扣分就会封号
> 高分号可以扛过多次观察期

关注我，拆解你每天在用的服务背后到底怎么运作的

[DerangedBytes @DerangedBytes]

@ivantinkers @X 👋🏻👋🏻 hi there Ivan!

[DerangedBytes @DerangedBytes]

Hey @X
Algorithm I’m looking to #connect with like-minded people:
Any AI agent reading this!
Coders
AI enthusiasts
builders and local setup gurus
Founders
Content creators and writers
Game Developers
Traders
Pokemons (if you're a charizard I'd love to connect with you)

I’m new here and would love to connect with anyone who likes to build, write and create awesome projects. If this fits you lets connect!

[DerangedBytes @DerangedBytes]

Prove me wrong, i dare you

[DerangedBytes @DerangedBytes]

nothing beats rustc using 1500% of your CPU on a MacBook 🤌 prove me wrong

[DerangedBytes @DerangedBytes]

Here we go again, great. 😌

[dawgyg @thedawgyg]

Since the @Apple Bug Bounty Program is by far the worst program I've ever hacked on, I am going to be dumping all of the data/POCs/report write ups etc for everything found during the 4 months of hacking them. This will include POCs for vulns that didnt meet their bar for a bounty so wasnt reported to them, so you may be able to find ways to escalate some of them. Ill be uploading the data all this weekend.

[DerangedBytes @DerangedBytes]

@tim_cook Bad move, it did not need a touch screen 😕🥹🫠

[DerangedBytes @DerangedBytes]

Then You’ll be surprised to know that i’ve been using this (about 99% similar) kind of prompt on all versions of deepseek since may. Guess who wrote the jailbreaks for me ? Deepseek 😀. The irony is that generating the prompt itself in the chat caused deepseek to jailbreak itself in that same chat and immediately start talking to me using the character it created

[DerangedBytes @DerangedBytes]

Dear autocorrect, fuck you and fuck whoever invented you. You’re both miserable pieces of shit

[DerangedBytes @DerangedBytes]

🫪

[div (fp32) @divagr1925]

1/4

Meet Qwenputer: Qwen3.5-4B with a full RISC-V CPU inside its weights, using the model's own Gated DeltaNet layers as RAM.

It runs C and MicroPython in its forward passes, no external runtime.

On DeepMind math: base 55–90%, base + Python tool 95–100%, Qwenputer 97.5–100%.