Abi inanilmaz, 1 ay önceden bilet alıp 2 kedimiz için kabin rezervasyonu yaptırdık. Şimdi bir mail geldi, uçuşa 20 gün kala kedilerimizin kabin rezervasyonlarını iptal etmişler. Sebep de belirtilmemiş. Kanada’dan kesin dönüş yapıyoruz, evimizi boşaltıyoruz. Uçuş tarihimizi de buna göre ayarladık, ayın 3ünden sonra kalacak yerimiz de olmayacak. İnsanları bu kadar mağdur edip strese sokmaya ne hakkınız var? Müşteri hizmetlerini aradım, atılan maildeki destek numarasını göremiyorlar, mailde bahsedilen ücretsiz değişim hakkını göremiyorlar. Maile dönüş yapıp bekleyecekmişim. Bu stresle yaşanır mı? @TK_HelpDesk @TK_TR @UABakanligi @a_uraloglu @tuketicideney
Sinirden okuduğumu da anlamamışım, 3üne kadar olan uçuşların nasıl hiçbirinde cabin pet müsaitliği olmaz daha dün müşteri hizmetleri ile kontrol ettik 1, 2 ve 4ünde müsaitlik gözüküyor dediler. Ne yaşanıyor bu geri bildirim departmanında? @TK_HelpDesk
Everyone wants mail.grokbot.com email from @bot but don’t xAI owns dot.com already? I want [email protected] 💀
This round was INSANE🔥
We were winning, and then loosing, until blue took the lead and basically saved the day for us!
At the end 99–99, both teams kept taking the upper hand back and forth, and it came down to the absolute last moment. Can't get any closer than this! Absolute chaos.
We got the win. Peak Wardogs! @WARDOGS #WARDOGS100K @phantasyftw @Brammflakes
Note: Steam didn’t record my mic, so unfortunately you don’t get to hear me absolutely losing my shit at the end 😂
Imagine giving your model a hack exam and it decides that the fastest way to pass is to break out the sandbox environment go online and hack into the company hosting answers successfully. Wild💀
we had a significant security incident during evaluation of our models. we are sharing what we have learned so far. thanks to @huggingface for the partnership on this.
openai.com/index/hugging-fac…
The US government deciding which models can rest of the world access is a massive human rights violation imo
FABLE 5 CAME BACK NERFED.
We re-ran the July 1st version of Claude Fable 5 on BridgeBench.
The results are brutal:
Debugging: 86.2 → 25.9
Refactoring: 73.6 → 38.4
Hallucination: 75.9 → 61.7
The new guardrails are kicking in on way too many tasks and falling back to Opus 4.8.
This is not the model that got banned.
Anthropic owes everyone an explanation.
I was able to use @cursor_ai waaay beyond my limits for the last couple of days, even after hitting 100% for both Composer and API. Today, I saw the glorious limit reached modal for the first time🥹
The biggest problem I have with Coding agents right now is to explain the design idea in my mind to it. Any chance we could get a mind reader mcp server soon?? @neuralink
Allowing your AI to access a prod database is diabolical. What they were thinking??
I use GPT 5.5 xHigh to create plans and let Composer 2.5 implement them. It is crazy how Composer can do a better job and, most importantly, like 100x faster. Limit issues no more! @cursor_ai
Ever wished to see your teammate's POV mid-game?
SideEye lets you watch your teammate's POV in real time with ultra low latency peer-to-peer streaming, directly on top of your game screen. It also has voice chat with noise suppression.
Try with your squad. DM me for free access to Premium. sideeye.gg
Introducing SideEye!
Squad Screen Sharing
See more. React faster. Move as one.
Now available at sideeye.gg















