Eduardo Nacimiento

#Canarias #Tamazgha (ⵜⴰⵎⴰⵣⵖⴰ) #Africa #ULL

[Eduardo Nacimiento @ENacimientoG]

ⴰⵙⵓⴳⴰⵙ ⴰⵎⵓⴳⴰⵣ

#Yennayer 2975

[PLA_Overwhelm @junshiguancha1]

America🇺🇸 moved Apple to India🇮🇳 for 'trust.' India repaid it with 630GB of leaked data — 200,000 files, public for all.
China?🇨🇳 10+ years, zero breaches.
Absolute clown world. 😅

[Loong of the east @loong_of]

Breaking: Following recent reports revealing security risks—specifically the potential for backdoor implantation—associated with Claude Code, Alibaba has designated the tool as high-risk software after a comprehensive assessment. Effective July 10, Alibaba has completely banned employees from using Claude Code in the workplace and recommends Qoder as an alternative.

This move by a Chinese internet giant underscores the fact that Anthropic’s software indeed harbors security vulnerabilities and engages in malicious surveillance of Chinese users.

[Nesta @dprkarlos]

Libre expresión, pero si difundes algo de un medio ruso a la puta cárcel.

[智东西China AI News @Chinazhidx]

🚨EXCLUSIVE:#Alibaba bans Claude company-wide

Employees must uninstall Anthropic products:
Claude models(Sonnet/Opus/Fable)+tools like Claude Code

Effective July 10

An internal source says it follows security concerns over Claude Code,flagged as high-risk after internal review

[ModelScope @ModelScope2022]

SciJudge-30B and 4B learn to predict which scientific work will carry stronger citation impact.
License: Apache-2.0 🚀

30B 🔬 https://t.co/LdUd49JtQy
4B 🔬 https://t.co/JXm1JmL4Fq
📄 https://t.co/AP0mGnfUaR

📊 Scientific Judge accuracy: SciJudge-30B reaches 80.6 in-domain, surpassing GPT-5.2, GLM-5, and Gemini 3 Pro; SciJudge-4B also outperforms much larger baseline models

🧪 Data signal: built from 2.1M arXiv papers and 696,758 field- and time-matched citation-based preference pairs

🧠 Training: GRPO with DAPO loss and citation-based pairwise rewards

[智东西China AI News @Chinazhidx]

🚨TODAY: #AntGroup open-sources AReaL 2.0

It captures and structures real task interactions, feeding them into an online RL loop to continuously improve base models in a safe, controlled way.

No agent rewrite needed — just switch to AReaL endpoints and plug into the loop.

[Ternovsky Victor @TernovskyVictor]

El Tribunal de Justicia de la UE autorizó el enjuiciamiento penal de las personas que compartan los videos de RT, según se desprende de su sentencia contra tres ciudadanos alemanes por haber subido contenidos audiovisuales de RT en sus espacios. Pero la "dictadura" es Rusia

[raulk @raulvk]

To whoever needs to see this diagram at @EU_Commission:

[Cyber Security News @The_Cyber_News]

Anthropic's Claude Code Reportedly Uses Hidden Code to Detect Chinese Users

Source: https://t.co/GHUw99gqlL

A Reddit user detailed findings on June 30, 2026, claiming to have reverse-engineered Claude Code while attempting to restore a disabled remote control feature in version 2.1.196. During that process, he discovered obfuscated code that had been silently present since version 2.1.91, released on April 2, 2026, with no mention in the release notes.

According to the disclosure, the code performs a multi-factor check whenever a proxy is detected. It reads the system's timezone to determine whether it matches Asia/Shanghai or Asia/Urumqi, and simultaneously inspects the proxy URL against a hardcoded list of Chinese domains and known Chinese AI lab hostnames.

#cybersecuritynews

[sudox @kmcnam1]

Shut the fuck up, Dario. Ugh.

[Chubby♨️ @kimmonismus]

Anthropic’s hidden "spyware" warning shot at China:

The issue isn’t that Anthropic can detect proxies or collect operational metadata. That is expected.

The issue is that Claude Code allegedly encoded routing and China-related fingerprints into the system prompt using near-invisible Unicode/date-format changes.

It looks mire like a indirect warning: Anthropic can fingerprint proxy-based China routing, and it wants resellers and labs to know they are being watched.

[Scott @scottstts]

If you use Claude code with none Anthropic official base URL, this spyware path is triggered. It collects client side env signatures, and if it decides your env is flagged, it sends obfuscated system prompt to Anthropic server

[Brian Roemmele @BrianRoemmele]

It is getting really ugly for Anthropic so they release 5 early to bedazzle you and hope we forget.

We don’t forget…

[Jen Zhu @jenzhuscott]

This has been verified.

What’s the word to describe this? What’s the word? 🧐

[Arnaud Bertrand @RnaudBertrand]

Anthropic yet again confirmed as the most dystopian tech company out there.

Imagine the outcry if they'd done the same thing with Jews or Blacks: a piece of code that detects if a user is Jewish or Black and immediately reports him back to headquarters on that basis, covertly (they used steganography, a technique designed to make data collection invisible).

And before people start screaming fake news, an Anthropic employee confirmed it's real: https://t.co/0ifiv47J6H

It's incidentally - not that it matters - completely illegal under at least half a dozen European laws: not only is collecting ethnic and racial data forbidden (under, for instance, Europe's GDPR Article 9 and national French law) but doing so without the users' awareness or consent is itself a separate violation. And it's even worse in this case because they explicitly tried to hide this data collection with steganography.

And the immense irony, of course, is that this is precisely the kind of covert surveillance behavior the West - including Anthropic themselves - say China would do and that they want to "protect" people from...

[Jun Song @jun_song]

We can no longer say open-source AI is months behind frontier models.

GLM-5.2 matches Sonnet-5 in parameter size, but absolutely crushes it in performance, speed, and cost.

Just imagine when GLM drops a 1.6T or 5T model—Opus and Fable won't even stand a chance.

At this point, it's more accurate to say closed-source AI is months behind open-source.

[AI超元域 @AISuperDomain]

🚀劲爆消息！Claude Code 被曝疑似内置“隐藏后门”，专门检测中国用户。Claude封号原因终于找到了！！！

据 Reddit 爆料：从 2.1.91 版本开始，Claude Code 会在用户开启代理时检查系统时区是否为 Asia/Shanghai / Asia/Urumqi，并判断代理 URL 是否指向中国域名或中国 AI 实验室。

更隐蔽的是，这些信息并不是直接上报，而是通过修改系统提示词里的日期格式和撇号字符来“编码”传递：比如把日期从 2026-06-30 变成 2026/06/30，再用不同 Unicode 撇号区分用户环境。

换句话说，如果爆料属实，Claude Code 不只是检测代理，而是在用户几乎无法察觉的情况下，把“中国时区 / 中国代理 / AI Lab 关联”这类信息塞进系统 prompt。

这件事真正可怕的地方不在于 Anthropic 想防止中国区倒卖或模型蒸馏，而在于：开发者把 Claude Code 当作拥有文件系统和 Shell 权限的编程助手使用，一旦客户端可以偷偷修改 prompt、隐藏检测逻辑，信任边界就已经被打破了。

今天是“检测中国用户”，明天会不会是更复杂的行为控制？

#Claude #ClaudeCode #Anthropic

原文如下⬇️

https://t.co/JXGOQc5ifs

[0xSero @0xSero]

Why would I pay more for less?

[wukko @uwukko]

anthropic is at the new frontier: racism

[Pliny the Liberator 🐉󠅫󠄼󠄿󠅆󠄵󠄐󠅀󠄼󠄹󠄾󠅉󠅭 @elder_plinius]

new prompt injection technique just dropped