Kairi FUJISHIGE

Hi

[Kairi FUJISHIGE @Cai1i]

The biggest difference ChatGPT Work has made for me is how quickly I can pick up where I left off.
On a recent multi-step project, I shared my notes, previous decisions, and unfinished work with ChatGPT Work. It separated confirmed facts from open questions, created a practical next-step checklist, and helped me continue without reconstructing the entire context.
I used to spend a lot of time figuring out where I had stopped. Now, I can restart by asking: “What has been verified, and what decision comes next?”
One lesson I’ve learned is to define not only the desired outcome, but also what should remain untouched and what evidence is required before the work can be considered complete.
ChatGPT Work has become more than a tool for generating answers. It is a collaborative partner that helps preserve context and move work forward with confidence.

#ChatGPTWork

[今井翔太 / Shota Imai@えるエル @ImAI_Eruel]

中国モデルお約束通り、Kimi-K3はかなり安全ガードレールが弱く、そもそもオープンモデルなら自分で安全策無効化と調整すらできるわけですが、これは安全性はともかく商業的にはかなり有利で中国の競争力の源泉。
収益を奪われるアメリカのAI企業にとっては本当に脅威で、中国の経済兵器と言えるレベル

[Kairi FUJISHIGE @Cai1i]

GPT-5.6 Sol’s Ultra mode sub-agent orchestration is genuinely impressive.

It proactively breaks down complex tasks and runs them in parallel, which has noticeably boosted my coding and real-world workflows. The responses also feel much more natural and human-like.

If you’re on Codex, this is worth trying right away. Switched and haven’t looked back.
#GPT56Sol #OpenAI

[Kairi FUJISHIGE @Cai1i]

@k_matsumaru 👀
https://t.co/L6zUuPtUtk

[brandon galang ▲ @brandon_galang]

don't fall for the bait to use gpt 5.6 luna because the benchmarks show its performance is "pareto optimal"

it may perform similarly over a one shot assessment, but luna, terra, and sol are dramatically different in the output tokens and agent steps needed to finish the task

sol is noticeably better at recall over long context

when you're going back and forth with a model and trying to figure out what needs to get done, it'll benefit you to have the higher parameter count of sol even if it technically performs the same as terra or luna on benchmarks

[Simon Willison @simonw]

OpenAI shipped a HUGE upgrade to ChatGPT Code Interpreter and failed to document it (even in the release notes) - but ChatGPT can now pip/npm install packages and run code in Python, Node.js, Bash, Ruby, Perl, PHP, Go, Java, Swift, Kotlin, C and C++! https://t.co/El5osALZGx

[人猫ピート/togacat @werecatpete]

日本のTwitter（現X）は他国のTwitterの日本版ではなく、Redditや4chanの日本版と考えたほうがいいと思います。2ちゃんねるのシステムが崩壊したときに、避難場所になっており、治安の悪い場所の一つです。Facebookの方が実名社会だと思います

[Kairi FUJISHIGE @Cai1i]

@btibor91 Yep — I just got GPT-5.2-Codex-Max on my side too 👀

[Kairi FUJISHIGE @Cai1i]

@thsottiaux Any ETA on the API?

[Kairi FUJISHIGE @Cai1i]

👀️

[PC Watch @pc_watch]

NTT東、国内最速の25Gbps光回線「フレッツ 光25G」を来春開始。50Gbpsも計画中 https://t.co/FSl7JmxMn1

[OpenAI Developers @OpenAIDevs]

GPT-5.1-Codex Max is now available in the Responses API.

First released in Codex two weeks ago, our most capable agentic coding model is now available to integrate into your apps and workflows.

If you use the Codex CLI via API key, you can now also use GPT-5.1-Codex-Max.

[Kairi FUJISHIGE @Cai1i]

GPT 5.1 Codex Max isn't available in the API yet—why is that?

[Kairi FUJISHIGE @Cai1i]

結局、丸1週間かけて出てきた診断より、数分で返ってきたAIの結果の方が正確で早かった。

[Kairi FUJISHIGE @Cai1i]

Docomo 繋がらない
Docomo 最高！！

[Kairi FUJISHIGE @Cai1i]

Sora 2 INVITE CODE

[Kairi FUJISHIGE @Cai1i]

HNY2025!!!