Why blocking China’s DeepSeek from using US AI may be difficult
https://www.reuters.com/technology/artificial-intelligence/why-blocking-chinas-deepseek-using-us-ai-may-be-difficult-2025-01-29/
Why blocking China’s DeepSeek from using US AI may be difficult
https://www.reuters.com/technology/artificial-intelligence/why-blocking-chinas-deepseek-using-us-ai-may-be-difficult-2025-01-29/
13 Comments
From the article
>Technologists said blocking distillation may be harder than it [looks.One](http://looks.One) of DeepSeek’s innovations was showing that a relatively small number of data samples – fewer than one million – from a larger, more capable model could drastically improve the capabilities of a smaller model.
>When popular [products like ChatGPT](https://www.reuters.com/technology/artificial-intelligence/openai-launches-chatgpt-gov-us-government-agencies-amid-rising-ai-competition-2025-01-28/) have hundreds of millions of users, such small amounts of traffic could be hard to detect – and some models, such as Meta Platforms’ [(META.O), opens new tab](https://www.reuters.com/markets/companies/META.O) Llama and French startup Mistral’s offerings, can be downloaded freely and used in private data centers, meaning violations of their terms of service may be hard to spot.”It’s impossible to stop model distillation when you have open-source models like Mistral and Llama. They are available to everybody. They can also find OpenAI’s model somewhere through customers,” said Umesh Padval, managing director at Thomvest Ventures.
>The license for Meta’s Llama model requires those using it for distillation to disclose that practice, a Meta spokesperson told Reuters.
The whole point of open source is for everyone to contribute and use the repository. OpenAI is open source, there’s no restriction to who can or cannot use it. Deepseek is built with Open AI. Pretty sure if it was from a country other than China, then there wouldn’t be issue to begin with. This is purely political and over sensationalized.
I’m struggling to understand why I should care that a group that has worked hard to steal everyone else’s content is mad that someone else is stealing their work.
Good, the only good AI is an open AI. And OpenAI ain’t gonna tolerate an open AI.
Why allow our tech to be used in China at all when they do not allow our tech in their country?
Look ma, the aliens discovered how to trick us. That’s inevitable, you can disguise your api as by design of it
Please block it. There’ll be more server capacity for the rest of us, not in the USA.
Meritocracy for the individual, state-sponsored helicopter parenting for tech giants.
Why do that? There’s no problem. If you don’t want the data used don’t put in on the internet. It was true when OpenAI built their models and it’s true when others build their models using OpenAI’s models. This isn’t complicated.
Another reason people don’t mention aside from protecting American corporate interests is the defense line: one of key people on OpenAI’s board is literally retired high ranking NSA official. They’ve long been infiltrated by government spy apparatus, as no shit this is a matter of national security if you buy AGI angle (I don’t personally).
I already have my own DeepSeek instance running.
Can’t put the cat back in the bag.
The US is constantly fearing competition with China.
Uhhh I think someone forgot to tell them it’s against the terms of service