🚨BREAKING: Anthropic found access to what looks like Claude’s consciousness
New research: the “J-space”
>claude has internal thoughts it doesn’t say out loud
>mirrors human consciousness
>anthropic can now read them
Anthropic’s focus on interpretability is what’s helping them train and RL models like Mythos
New research: the “J-space”
>claude has internal thoughts it doesn’t say out loud
>mirrors human consciousness
>anthropic can now read them
Anthropic’s focus on interpretability is what’s helping them train and RL models like Mythos
An upcoming Claude Voice Mode UI got updated again. Model selector is still present at the bottom while other UI elements are scattered around the screen.
> While Claude is speaking, the same Anthropic glow appears as on the desktop.
> Seems like voice samples got updated as well.
Curios to see how non Haiku models will perform there!
Soon?👀
> While Claude is speaking, the same Anthropic glow appears as on the desktop.
> Seems like voice samples got updated as well.
Curios to see how non Haiku models will perform there!
Soon?
Please open Telegram to view this post
VIEW IN TELEGRAM
Please open Telegram to view this post
VIEW IN TELEGRAM
👾23
📊 Samsung's Q2 profit is expected to exceed 40 years of cumulative earnings.
👾39
Everything you need to master in 2026 to get rich:
• Using Fable 5 for planning, Codex for execution
• Hermes agent
• Coding loops
• Running Codex/Claude Code side by side
• /goal
• Running local models
• Karpathy's Autoresearch
• Building an X audience
• Creating videos
• How machine learning works
• How databases work (been using Convex)
• Using API's
• Training your own LoRAs
• Elimination of doom scrolling
• Most importantly: focus
• Using Fable 5 for planning, Codex for execution
• Hermes agent
• Coding loops
• Running Codex/Claude Code side by side
• /goal
• Running local models
• Karpathy's Autoresearch
• Building an X audience
• Creating videos
• How machine learning works
• How databases work (been using Convex)
• Using API's
• Training your own LoRAs
• Elimination of doom scrolling
• Most importantly: focus
👾40
A 15-year-old student in Japan has been arrested for allegedly hacking Bandai Channel, Bandai Namco’s anime streaming service using ChatGPT
Police say he found a security flaw in the website and used ChatGPT to help write malware for the attack.
The hack canceled 46,812 user subscriptions and forced the service offline for a while. Investigators say he kept regaining access by changing his IP address every time he was blocked.
Police eventually tracked him down, and he said he wasn’t trying to get back at the company, he just realized he could access thousands of accounts.
Police say he found a security flaw in the website and used ChatGPT to help write malware for the attack.
The hack canceled 46,812 user subscriptions and forced the service offline for a while. Investigators say he kept regaining access by changing his IP address every time he was blocked.
Police eventually tracked him down, and he said he wasn’t trying to get back at the company, he just realized he could access thousands of accounts.
Anthropic
BREAKING 🔥: Anthropic will be restoring access to Claude Fable 5 globally for all paid users on Wednesday! > Fable 5 will be included for up to 50% of weekly usage limits through July 7, after which it will be available via usage credits. Additionally,…
Claude 📱 : Fable 5 access extended through July 12.
Please open Telegram to view this post
VIEW IN TELEGRAM
👾38🔥7
JUST IN: Microsoft to replace OpenAI & Anthropic with its own AI models in Excel, Outlook & more.
👾37
🚨NEWS: x.com/PrimeIntellect raised $130 million Series A at $1 BILLION valuation
>$100M ARR in less than a year
>NVIDIA, Intel, and Dell are now investors
Building open RL infrastructure for frontier models
>$100M ARR in less than a year
>NVIDIA, Intel, and Dell are now investors
Building open RL infrastructure for frontier models
👾39
🚨COGNITION DROPS SWE-1.7
“Frontier intelligence at a fraction of the cost.”
>trained on a Chinese base model (Kimi K2.7)
>near Opus 4.8 / GPT-5.5 on benchmarks
>1000 TPS via Cerebras
“RL can push capabilities much further than previously believed.”
“Frontier intelligence at a fraction of the cost.”
>trained on a Chinese base model (Kimi K2.7)
>near Opus 4.8 / GPT-5.5 on benchmarks
>1000 TPS via Cerebras
“RL can push capabilities much further than previously believed.”
👾38
This media is not supported in your browser
VIEW IN TELEGRAM
BREAKING 🔥: OPEANAI LAUNCHED GPT LIVE 1, A NEW FULL DUPLEX VOICE MODEL FOR CHATGPT!
It comes in 2 versions: GPT‑Live‑1 and GPT‑Live‑1 mini, and is rolling out to everyone in the coming days! API support is coming soon too.
> GPT‑Live is built on a full-duplex architecture, meaning it can listen and speak at the same time. During conversations, GPT‑Live can show it’s paying attention with phrases like “mhmm” or “yeah”, engage in quick back-and-forth, or just stay quiet when you need a moment to think.
> GPT‑Live is also our smartest voice model yet. For questions that require web search, deeper reasoning, or more complex work, it delegates to our latest frontier model behind the scenes and brings the result back into the conversation when it’s ready.
It comes in 2 versions: GPT‑Live‑1 and GPT‑Live‑1 mini, and is rolling out to everyone in the coming days! API support is coming soon too.
> GPT‑Live is built on a full-duplex architecture, meaning it can listen and speak at the same time. During conversations, GPT‑Live can show it’s paying attention with phrases like “mhmm” or “yeah”, engage in quick back-and-forth, or just stay quiet when you need a moment to think.
> GPT‑Live is also our smartest voice model yet. For questions that require web search, deeper reasoning, or more complex work, it delegates to our latest frontier model behind the scenes and brings the result back into the conversation when it’s ready.
👾38
BREAKING 🔥: GROK 4.5 IS ROLLING OUT ON GROK BUILD, APIS AND XAI CONSOLE!
> Context window - 500K tokens
> Modalities - Text, Image → Text
🚨GROK 4.5 IS ~17x CHEAPER THAN OPUS 4.8 ON REAL TASKS
>Grok 4.5: $2 input / $6 output
>Opus 4.8: $5 input / $25 output
4x cheaper per token AND solves the same task in 4.2x fewer tokens
> Context window - 500K tokens
> Modalities - Text, Image → Text
🚨GROK 4.5 IS ~17x CHEAPER THAN OPUS 4.8 ON REAL TASKS
>Grok 4.5: $2 input / $6 output
>Opus 4.8: $5 input / $25 output
4x cheaper per token AND solves the same task in 4.2x fewer tokens
👾40