Re: DeepSeek

On Wed, 29 Jan 2025 21:43:21 +0000, Physfitfreak wrote:

On 1/28/25 11:02 AM, Physfitfreak wrote:

On 1/27/25 11:39 PM, The Starmaker wrote:

In the Netflix doc you can see the openai guy [Brockman] go to his
computer and realizes he's in 'user mode' and you can see him switch
to...Godmode
on the computer screen.

>
>
Well, when China steps in, all of them are sorry asses.
>
OpenAI people murder those who criticize them also. They killed one of
their own employees a month ago only three months after he quit to blow
the whistle for their illegal activities.
>
Science/tech is in the wrong hands.

>
>
China has released Qwen also, which is said to be superior to all the
four other AI including DeepSeek and Claude.
>
Try it at:
>
https://chat.qwenlm.ai/

I tested it with the theory about Mercury and the deflection of
starlight by the Sun. I compared the answers with those of DeepSeek and
ChatGPT.
It proved that qwen is FULL OF SHIT, making mistake after mistake, and
WRITING SHIT about the formulae, mixing the theory of Nov. 18, 1915 with
Schwarzschild, Hilbert (1917) and the inclusion of the cosmological
constant.
Once I contested its replies, it agreed and consolidated his answers in
a shameful way. It's A VERY LAME AI ENGINE. Maybe useful for finance or
software, but no more than that.
Also, it LIES OPENLY AND WITHOUT SHAME. Fuck the site.