MB

unrealmarc

AI & ML interests

None yet

Recent Activity

liked a Space 6 days ago

webml-community/TinySwallow-1.5B-Instruct-WebGPU

upvoted an article 6 days ago

G2P Shrinks Speech Models

liked a dataset 8 days ago

cais/hle

View all activity

Organizations

unrealmarc's activity

liked a Space 6 days ago

TinySwallow-1.5B-Instruct-WebGPU

🐦

A compact Japanese LLM that runs locally in your browser.

upvoted an article 6 days ago

Article

G2P Shrinks Speech Models

•

6 days ago

• 23

liked a dataset 8 days ago

cais/hle

Viewer • Updated about 9 hours ago • 2.7k • 4.43k • 215

liked a model 15 days ago

Qwen/Qwen2.5-VL-3B-Instruct

Image-Text-to-Text • Updated 8 days ago • 182k • 184

reacted to hexgrad's post with ❤️ 17 days ago

Post

3910

IMHO, being able & willing to defeat CAPTCHA, hCaptcha, or any other reasoning puzzle is a must-have for any Web-Browsing / Computer-Using Agent (WB/CUA).

I realize it subverts the purpose of CAPTCHA, but I do not think you can claim to be building AGI/agents without smoothly passing humanity checks. It would be like getting in a self-driving car that requires human intervention over speed bumps. Claiming AGI or even "somewhat powerful AI" seems hollow if you are halted by a mere CAPTCHA.

I imagine OpenAI's Operator is *able* but *not willing* to defeat CAPTCHA. Like their non-profit status, I expect that policy to evolve over time—and if not, rival agent-builders will attack that opening to offer a better product.