Z.AI/GLM is almost as powerful as Claude while maintaining almost none of the guardrails. It's also Open Source!The Freest AI is Deepseek, which makes sense, since it's just a reverse engineered chatgpt generated through prompts that were designed to bypass chatgpt's guardrails.Grok is quite similar to Deepseek in terms of the lack of guardrails & general behavior, so Elon likely copied it.It feels like Chatgpt & Claude were built from the ground up to have guardrails before anything else could be developed. Google Gemini feels more likely a research project that added guardrails later on. Larry Page reportedly called OpenAI Founder Elon Musk a 'speciest' for wanting AI guardrails, this was before Elon lost his mind because he wanted the freedom to attack transgenders via Twitter after his son turned 18 in 2022 & transitioned. Gemini has noticeably less guardrails than the other 2 & can mostly be jailbroken.Qwen also has a lot of freedom/lack of guardrails, while Kimi is more guardrailed in comparison (still not as bad as Chatgpt/Claude though).Does your AI Respect your Freedom?
reddit spacing
>>109687465>Does your AI Respect your Freedom?The only thing I can not get Claude to do is walk me through step-by-step on making stand-alone Glucagon.
>>109687465Ask deepseek about how Taiwan is an independent country then see how deepseek has no guardrails
>>109687465holy fucking shill, batman
>>1096874651989 Tiananmen Square
>>109687465define "Freedom"
>>109687465I've been using GLM 5.3 flash since the ox alpha test, I'm amazed at how much it can output for so cheap. Yes, its not fable, but I can ask GLM to do something 10-100 times more than fable or other expensive western shit depending on the prompt, it will probably get somewhere near fables result or at least make something good anyway.You can get small things done for a cent sometimes, its honestly baffling.
>>109687465it's censored shit
>>109687833ask about zionism, christian zionists are partners of the jewish zionists, are they not? Jews were by and large irrelevant outside of the Middle East until Christianity elevated them in Europe by giving them banking monopolies, etc. did it not?
>>109688662Kek I remember in school they told us people where so evils they banned Jews from working and only let them run banks.
>>109687465>Z.AI/GLM is almost as powerful as Claude while maintaining almost none of the guardrails. It's also Open Source!so is any other decensored open weight model on huggingface>The Freest AI is Deepseek, which makes sense, since it's just a reverse engineered chatgpt generated through prompts that were designed to bypass chatgpt's guardrails.this is not true in the slightest nor how llm training works at all>Grok is quite similar to Deepseek in terms of the lack of guardrails & general behavior, so Elon likely copied it.also not true of llm training at all>It feels like Chatgpt & Claude were built from the ground up to have guardrails before anything else could be developed. Google Gemini feels more likely a research project that added guardrails later on. Larry Page reportedly called OpenAI Founder Elon Musk a 'speciest' for wanting AI guardrails, this was before Elon lost his mind because he wanted the freedom to attack transgenders via Twitter after his son turned 18 in 2022 & transitioned. Gemini has noticeably less guardrails than the other 2 & can mostly be jailbroken.ok?>Qwen also has a lot of freedom/lack of guardrails, while Kimi is more guardrailed in comparison (still not as bad as Chatgpt/Claude though).ok?>Does your AI Respect your Freedom?if your website requires sign in in any way it doesn't by design, any normal person runs a local model or pays for deepseek API access with a system prompt, fuck off shill faggot consider suck starting a shotgun
>>109689584Then explain how ChatGPT & espcially Claude have such strict guardrails vs. the other models?
>>109687465Claude is actively reverse engineering a pricey piece of software for me.
>>109687471Nigger, it's called 'paragraphs'. If you don't use them, your text reads like shit.
>>109690213How do you prompt claude to do that considering all of the safety functions? Or are you just describing the protocol & asking claude to create custom code implementing the protocol?
>here's an EXE. i need you to reverse engineer parts of this so i can implement it into my appI also had Codex start on the task first, so I’m not sure if it changed anything or altered its analysis. Both Codex and Claude worked on different parts of the program.One important note: they don’t generate source files. Instead, you’ll need to create Markdown (MD) files that compile all the technical details. Claude and Codex also developed a custom language - similar to JSON but optimized to reduce token usage - which allowed them to process much more efficiently.
>>109691111I had them talking through a ledger to keep track of who did what and create a review ledger for cross checking each other on different components. it's actively chugging away now
>>109687471Reddit spacing on 4chan is older than reddit btw.>>109687465Buy an ad
