Umut

Security Systems Expert | AI Researcher

[Umut @UmutTuzmen]

Sunday morning.

Sitting outside, watching nature.

Coffee in one hand.
Phone in the other.

Old habits die hard. 📱☕

[Umut @UmutTuzmen]

@dylanmccavitt @ClaudeDevs Perfect timing. 😂

My weekly limit reset 2 hours ago, and now you reset everyone again.

Thanks for resetting my freshly reset limit.

[Umut @UmutTuzmen]

@ClaudeDevs If you reset the weekly limits, we can test your suggestion 😆

[Umut @UmutTuzmen]

@claudeai Fable used up my weekly limit incredibly fast. I’ve already waited two days for the reset, and still have three more to go. At this point, the limits are blocking real work more than helping productivity. @ClaudeDevs #Claude @bcherny

[Umut @UmutTuzmen]

@_catwu We’ll build something… if the Fable 5 limits let us. 😣
The model is great, but the limit disappears way too fast.
One serious session is enough to hit the Fable 5 limit, and the weekly limits disappear just as fast.

[Umut @UmutTuzmen]

@claudeai Tek bir ciddi oyun seansı bile Fable 5 limitine ulaşmak için yeterli ve haftalık limitler de aynı hızla tükeniyor.

Harika model. Sinir bozucu limitler. @ClaudeDevs @bcherny #claude #claudeai

[Umut @UmutTuzmen]

@elonmusk If you exploit and bomb their countries, they will conquer you from within.

[Umut @UmutTuzmen]

Something is wrong. #claude @claudeai @ClaudeDevs

[Umut @UmutTuzmen]

@meyusufdemirci @Kerem_Erkan En genç 12 saat olmak üzere 4-5 revize ve akabinde itiraz cevabıyla hemen onayladılar.

[Umut @UmutTuzmen]

@xMysteryLC @rillau144 😆😆

[Umut @UmutTuzmen]

@atillayurtseven Workflow

[Umut @UmutTuzmen]

@claudeai No Fable 5, no subscription.

Simple as that.

[Umut @UmutTuzmen]

@TrollFootball 😆😆

[Umut @UmutTuzmen]

@AnthropicAI It has been banned.

[Umut @UmutTuzmen]

13 minutes.

103 agents.
3.4M tokens.

We're rapidly approaching the point where asking a question feels less like using software and more like hiring a temporary research firm. @claudeai #claude #opus 😳

[Umut @UmutTuzmen]

Me: Create a PDF.
Fable 5: This may involve dangerous cybersecurity and biology topics.
Me: It's a PDF...
@claudeai @ClaudeDevs #claude #fable5

[Umut @UmutTuzmen]

What do you think about the error below?

"Fable 5's safety measures flagged this message. They may also flag safe, normal content. These measures allow us to offer Mythos-level abilities earlier in other areas, and we're working to improve them. Upgrade to Opus 4.8. Send feedback or learn more."

[Umut @UmutTuzmen]

@ClaudeDevs Tried Claude Fable 5 today. Mid-session it downgraded itself to Opus 4.8 because of a safety flag.

Didn't expect that. Power with built-in guardrails. Interesting call @claudeai 🤔

[Umut @UmutTuzmen]

Claude Fable 5 just dropped in Claude Code.

Mythos-class model, now available for general use. Better code quality, less prompting, longer sessions.

Heads up though — it burns through tokens 2x faster than Opus 4.8.

Going to test it on my own projects today. Will report back.

[Claude @claudeai]

Introducing Claude Fable 5: a Mythos-class model that we’ve made safe for general use.

Its capabilities exceed those of any model we’ve ever made generally available.

[Umut @UmutTuzmen]

I've spent years installing security systems. I know what good sensors look like. But watching Tesla FSD navigate a narrow mountain road in China, steep cliffs on both sides, cameras only, no lidar, zero mistakes.

If your software is good enough, the hardware stops being the limiting factor.
Good Job Elon 👏

[Umut @UmutTuzmen]

Anthropic just published something worth reading carefully.

Claude now writes more than 80% of their code. Engineers ship 8x more work than 2 years ago. Claude handles 12-hour tasks alone.

Next step? AI training and designing its own successor. They call it recursive self-improvement.

Not there yet. But clearly on the way.

This isn't just a tech update. It might be one of the most important turning points in human history.

https://t.co/ANrLalyn4i