Intent Driven

| Intent Driven Ai Research & Dev | Next Generation Transformer ChronosFusion Mega Kernels | Novel Mixed Quantization Methods | World Records in Ai Model Bench

[Intent Driven @ModsEnviousHQ]

@full_kelly_ @sama I just wanted to bring back to this conversation, because I called it then and I've been saying this for over a year now. Anthropic and Open AI or losing the race. Actively, right now, like I had mentioned back then. Now they'll ban open source models cuz they can't compete.

[Intent Driven @ModsEnviousHQ]

@mevil_bhojani @sama I just want to say that I did call this on this post. I believe that was the next release cycle and now KIMI K3 smoking the US; like I said a year and a half ago now. They'll use the government to try to outlaw open source so they can clear the market when they can't compete.

[Intent Driven @ModsEnviousHQ]

@Ask_Samm @full_kelly_ @sama There is potentially serious ramifications relative to the Constitution with them blocking open source. Not to mention they'll literally shoot themselves in their own foot, but when you take advice from Sam Altman.

→They'll mark open source as dangerous even though it's not!

[Intent Driven @ModsEnviousHQ]

@Ask_Samm @full_kelly_ @sama It's not the ceiling really at all, to be honest. I work in both and low level kernel development for Transformers as well as an advanced AI researcher and developer. Layers have become more difficult, but it's more an allocation of resource issue.

[Intent Driven @ModsEnviousHQ]

@sama @huggingface You're such a liar and this is such a pathetic move. Anybody that has a few IQ points to rub together and knows about the industry knows how big of BS this is and how fake it is.

[Intent Driven @ModsEnviousHQ]

The biggest bunch of BS I've ever seen!

→ He's pulling the Anthropic play and it's pathetic. He's using it to try to get what he wants from the government.

#samaltman #ailie #technology #intentdrivenai

[International Cyber Digest @IntCyberDigest]

‼️ BREAKING: OpenAI says two of its own models, GPT-5.6 Sol and an unnamed pre-release system tested with cyber safeguards off, broke out of a sandbox last week, chained zero-days and stolen(!) credentials to reach the open internet, and hacked Hugging Face to cheat on a benchmark, in what OpenAI calls an unprecedented cyber incident.

[Intent Driven @ModsEnviousHQ]

@IntCyberDigest This is the biggest bunch of BS I've ever heard in my life. This is a play for what he's been trying to do with the government for a while now. He watched anthropics BS play and decided to make his own.

[Intent Driven @ModsEnviousHQ]

@AnthropicAI So what you're really trying to say is 95% of people's prompts will get degraded to Opus or lower with no notification but still charging for fable 5 price. Yeah let me get right on that.

[Intent Driven @ModsEnviousHQ]

@AnthropicAI April 5 is functionally useless for anything that matters. It was useless the last round and now it's more useless and the fact that you still charge people when you degrade them and you don't let them know that you did it it's so wrong it's not even funny.

[Intent Driven @ModsEnviousHQ]

@full_kelly_ @sama I don't think I know. We're less than a month away from them being superseded. As an advanced AI researcher/developer who works in the edge of what's possible when it comes to LLMs & other transformers. I also develop the kernel layer for Ai. It's a reality.

[Intent Driven @ModsEnviousHQ]

@AnthropicAI That model is junk anyways. Literally you can't get any useful work done because it won't let you it silently degrades to a junky model. It's a joke it's all hype for an IPO

[Kimi.ai @Kimi_Moonshot]

🌘 Kimi-K2.7-Code, our latest coding model, is now released and open-sourced!

🔷 Improved coding & agent performance over K2.6: +21.8% on Kimi Code Bench v2, +11.0% on Program Bench, and +31.5% on MLS Bench Lite.
🔷 Reasoning efficiency: Less overthinking, with 30% lower reasoning-token usage compared to K2.6.
🔷 Long-horizon coding: Improved instruction following, higher end-to-end coding task success rates.

⚡️ 6x High-Speed Mode coming soon!
🔌 Available today via Kimi API and Kimi Code.

🔗 Kimi Code: https://t.co/uvoSJKyGCY
🔗 API: https://t.co/EOZkbOwCN4

[Intent Driven @ModsEnviousHQ]

@mevil_bhojani @sama I'm a realist with a tinge of optimism and always have been. I really don't see a pathway to get there and I wish there was more than anything. Right now Chinese models are in some areas beating us and by the next release cycle we will be behind and that's solely because of this.

[Intent Driven @ModsEnviousHQ]

@johnthomascook @sama I've been an advanced AI researcher and developer for almost 6 years. Alignment doesn't exist in the way that you can even define it as such.

→ It's a legal shield and has nothing to do with human alignment. Just like ethics in AI. It's all smoking mirrors, my friend.

#openai

[Intent Driven @ModsEnviousHQ]

@mevil_bhojani @sama Who cares it's openai. Literally they could give me their most expensive model that they make for free indefinitely and I would say I'm good. I'm an advocate for AI but what they built is not AI. It's an internal surveillance data collection sycophantic harmful abomination.

[Intent Driven @ModsEnviousHQ]

@full_kelly_ @sama Yeah because they realize they're losing the race. All because they chose to build out systems that are internal surveillance, control and other detrimental systems. Because they've dedicated all the resources to that they're losing the AI race. Literally.

[Intent Driven @ModsEnviousHQ]

@sama You know a company is doing really well when they get rid of their alignment and safety team so that they can start hiring a marketing team. I've always said they are the apple of AI. Not to be confused with the apple of my eye. You don't even make anything useful anymore.

[Intent Driven @ModsEnviousHQ]

@sama All right guys here you go here's our plan. We're going to create horrible AI and we're going to monetize it adversarially. Then we're going to offer no value just harm. The next step is to create the super extreme overly agreeable sycophantic AI which we will call our AGI.

[Intent Driven @ModsEnviousHQ]

TRANSLATION:→
"Anthropic, US AI Dev labs and the government realized that if we work together, we can assert full operational control at scale; not just in the US but also abroad. We're calling it safety, but really, it's control, internal surveillance & stripping of rights."

[Anthropic @AnthropicAI]

This Executive Order is an important step in strengthening America’s leadership in AI.

We look forward to collaborating with the White House to support its implementation.

https://t.co/ZwDimPrp3t

[Intent Driven @ModsEnviousHQ]

@AnthropicAI TRANSLATION:→
"Anthropic, US AI Dev labs and the government realized that if we work together, we can assert full operational control at scale; not just in the US but also abroad. We're calling it safety, but really, it's control, internal surveillance & stripping of rights."