Juhyung Park retweeted
In lieu of flowers, memorial contributions may be made to:
Johns Hopkins Alzheimer’s Disease Research Center Fund
alzresearch.org/giving/
Alzheimer’s Foundation of America Caregiver Resources
alzfdn.org/donate/
Photo Credit: Tommy Garcia
Juhyung Park retweeted
제목만봤음에도 웃음을참을수없었음... 너희가 고생이 많다 그래 blog.cloudflare.com/k-isms/
Juhyung Park retweeted
【PHPセキュリティ問題 時系列順まとめ】
今回のPHPの事案を時系列順にまとめてみました。
----
2024年:PHPの潜在的な脆弱性を発見
外部監査をきっかけにメモリ管理の問題を認識。
しかし、修正方針がまとまらず約2年間対応が停滞。
↓
2026年8月:重大な任意コード実行リスクが判明
Calif(セキュリティ企業)やOpenAIなどが、認証なしで任意コード実行(RCE)が可能な攻撃経路を報告。
PHP開発チームが修正を最優先で進める。
↓
10月21日 0時頃:脆弱性情報が事前流出
修正パッチ公開前に、2件のCVE番号がHacker Newsへ投稿され、←何してんの
それがSNSや掲示板にも拡散。
※この時点では技術的な詳細は含まれていなかった。
↓
10月21日 4時30分:PHP公式が緊急告知
情報流出を受け、パッチ公開予定を22日から21日23時(日本時間)へ前倒し。
↓
10月21日 5時頃:攻撃実証動画が誤公開
Califが攻撃の動作動画を誤って公開。←最高にカス、バカ
すぐに削除されたものの複製が拡散し、攻撃コードの開発が加速。
↓
10月21日:研究者らによる未検証情報が拡散
パッチ公開前に、@ J4ck3LSyN などによって脆弱性の原因や再現方法をSNSなどで発信する動きが発生。←余計なことをするな
↓
10月21日 16時41分:修正済みバイナリを先行公開
Chainguard(セキュリティを専門とする米国企業)が公開予定の約6時間前に修正済みコンテナイメージを配布。←最大最悪のカス、超絶ルール違反
これにより攻撃者によるパッチの差分解析が可能な状態に。
※PHP Foundationもこれを一方的なエンバーゴ違反(事前に取り決めた情報の公開禁止期間を守らないこと)と明言しています。
↓
10月21日:国内外でサービスの緊急停止が相次ぐ
修正パッチがまだ提供されていないため、国内外のWebサービスや政府機関が予防措置として一時停止・メンテナンスを実施。
↓
10月21日 23時頃:PHP公式が修正版を公開
CVE-2026-109618、CVE-2026-109619への修正を含むPHP 8.2〜8.5の更新版を公開。
↓
10月22日:情報流出と管理体制の問題が焦点に
PHP Foundationが、2024年からの修正停滞を組織的な失敗と認める。協調的な脆弱性開示が崩れた経緯も問題視される。
----
【結論】人災。
特に、
⚠️ Califが攻撃の動作動画を誤って公開したこと
⚠️ Chainguardの勝手な判断による修正済みバイナリを含むコンテナイメージの公開
が最悪。終わってます。
Effective Immediately, all organizations using PHP for their systems are advised to be prepared for an update addressing two critical vulnerabilities. PHP is widely used for web applications. These vulnerabilities often provide a direct path to the server's file system or database. If an immediate update is not feasible, it is recommend that organizations take the system offline until it can be fully mitigated. NSA will follow up shortly with more specific guidance.
Every Korean bank (probably by law) forces you to install system-wide spyware apps that make you log in with an on screen keyboard that you have to click the letters of your password with your mouse.
Just in case you want to know about the talent at these banks.
Last week some of South Korea's biggest banks were hit by a cyberattack. Thanks to a report from CrowdStrike tonight, we now know the entire hack may have been done by a single person. He used a combined stack of an open-source AI penetration tool named ARTEX, DeepSeek v4.1-Flash, GLM-5.3, Grok 4.6, and Claude Code.
Juhyung Park retweeted
⚡️ Linux Kernel just issued a single-fix security update for stable LTS branches, addressing CVE-2026-64560.
We were able to exploit the bug and independently confirm that it's serious.
The impact is kernel code execution and EoP from unpriveleged user with no CAPs/userns, io_uring disable + BPF JIT hardening. The use-after-free primitive is reliable despite the race (<1m to exploit).
No wild attacks were seen yet. They are likely, given our technical picture.
Majority of Android devices are affected. Exploitability on Android wasn't empirically confirmed yet.
Juhyung Park retweeted
Run Kimi K3 on a Mac Studio 🫰
K3 is 2.8T parameters and 1.6TB on disk, which makes it impossible to run on Apple Silicon.
Until now.
Our MLX port is now open source: github.com/PipeNetwork/kimi-…
To accomplish this, we solved two things:
1. We wrote a streaming converter that walks one layer at a time, so that mlx_lm doesn't need to materialize the whole model.
2. REAP pruning sits on top and scores all 896 experts against a calibration corpus to keep only ones your workload needs.
That's what brings K3 down to 350GB and inside a Mac Studio.
Releasing the model weights and technical report of Kimi K3.
Kimi K3 is our most capable model: a 2.8T MoE model with native visual understanding and a 1M-token context window.
New model architecture: 2.5x the intelligence per unit of compute, not just more params.
Alongside Kimi K3, we're opening up more of the stack behind it — high-performance attention kernels, MoE communication library, and infrastructure for running agent environments at scale.
Model weights: huggingface.co/moonshotai/Ki…
Tech report: github.com/MoonshotAI/Kimi-K…
Tech blog: kimi.com/blog/kimi-k3
Juhyung Park retweeted
GPT-5.6 tops the DeepSWE leaderboard at 73%.
Sol, Terra, and Luna results are now available.
When Fable 5 is used for frontier LLM development, it does not notify the user and instead limits the model’s capabilities through methods such as prompt modification, steering vectors, and PEFT.
Anthropic estimated that this would affect approximately 0.03% of traffic.
Juhyung Park retweeted
mythos will be bad ON PURPOSE on ai "frontier llm research" tasks, this is very very sad for the research community
also the fact that this is un purpose not visible to the user is crazy
Juhyung Park retweeted
Replying to @arter97
I ran it in Windows Sandbox and it did delete things to the point the sandbox is unresponsive (though a lot of errors due to permissions). Did have to add --no-preserve-root
So now we have a universal rm -rf / --no-preserve-root
Juhyung Park retweeted
If low-level systems excite you, then you will enjoy reading this branch prediction blog from cloudflare.
blog.cloudflare.com/branch-p…






































