Claude just released

@startupsep

6.2K followers · usually 1.3K views a post

Views

5.6K

Likes

175

Comments

16

Shares

19

Selling · SaaS › AI Tool

Claudeseen in 998 videos →

Why it worked

The video taps into the current hype around AI advancements while also raising a critical concern about accessibility and potential societal divides, making it a thought-provoking and shareable piece of content.

Summary

The video discusses the release of Anthropic's latest AI model, Claude, highlighting its impressive performance but also its inaccessibility due to significant cybersecurity risks. The creator expresses concern that this could lead to a future where only the wealthy have access to the most advanced AI.

Structure

1Introduction of Claude's new model
2Discussion of its impressive benchmarks
3Explanation of inaccessibility due to cybersecurity risks
4Concerns about a future AI divide
5Mention of Google's Gemma as an alternative

Product placement

Claude: an AI model that generates text. It ACTS as the subject of the video. Removing it would change the video's topic. Claude Opus 4.6: an AI model that generates text. It ACTS as the subject of the video. Removing it would change the video's topic. Mythos Preview: an AI model that generates text. It ACTS as the subject of the video. Removing it would change the video's topic. Gemma: an AI model that generates text. It ACTS as the subject of the video. Removing it would change the video's topic. Google: a company that develops AI models. It merely APPEARS as the developer of Gemma. Removing it would not change what happens in the video.

Original caption

Claude Mythos preview is OUT! kinda… #mythos #claude #ai

More from @startupsep

Their other posts in the index, biggest breakout first.

On-screen text

Project Glasswing Securing critical software for the AI era SWE-bench Pro Mythos Preview 77.8% Opus 4.6 53.4% Terminal-Bench 2.0 Mythos Preview 82.0% Opus 4.6 65.4% SWE-bench Multimodal (internal implementation) Mythos Preview 59.5% Opus 4.6 53.3% SWE-bench Multilingual Mythos Preview 77.3% Opus 4.6 66.6% Mythos Preview and our next-best model, Claude Opus 4.6: Cybersecurity Vulnerability Reproduction CyberGym Mythos Preview Opus 4.6 66.6% In addition to our partners have a of our Claude Mythos Preview they've found: This is what THE FIRST TIME WHERE ME OR OR THE GENERAL CONSUMER CAN'T USE THE BEST MODEL NOW THIS RELEASE BECAUSE UP TO ALL OF US SHIP AT THE BE USING THE THERE. BUT NOW I WORLD RICHEST PEOPLE WILL HAVE THE ACCESS TO THESE AMAZING AND THE REST AT THE SAME GOT GEMMA RELEASED BY GOOGLE, OF THE BEST SO I DON'T KNOW, IT COULD WHAT YOU THINK? LATEST AI.

Transcript

Claude just released their latest model, that is the model, that humanity has ever created. Except you can't access it. Now, before I get into that, this model is absolutely insane. The benchmarks that they released is just crushing Opus 4.6, which was already insane. But I'm pretty sad about this release. The first time where me or you or the general consumer can't use the best model that's out there. Now, for good reason. Anthropic says that this model poses such an insane cyber security threat that it's already found a bunch of security vulnerabilities on major apps that you're already using today. And if they put this in the hand of general consumers, we will all be doomed because everything would start getting hacked. So they've only released it as a preview to some of these big companies who can start using it for cyber security purposes. This makes sense, but I'm sad because up to this point, all of us were able to ship at the same speed, be using the best models out there. But now I see a potential world where only the richest people will have the access to these amazing models and the rest at the same time, got Gemma released by Google, which is one of the best open source model. So I don't know, it could go either way. What do you think? Latest AI.