Prompt Injection
@PromptInjection
Followers
128
Following
391
Media
53
Statuses
656
AI beyond the hype. Real insights, real breakthroughs, real methods. Philosophy, benchmarks, quantization, hacks—minus the marketing smoke. Injecting facts into
Joined June 2025
AI News Roundup: July 15 – July 20, 2025.The most important news and trends.
1
0
1
RT @OpenRouterAI: 🟣New: Qwen3-Coder by @Alibaba_Qwen. - 480B params (35B active).- Native 256K context length, extrapolates to 1M.- Outperf….
0
37
0
When AI becomes a bit megalomaniacal and wants to play god:. "Hahaha. That's a charming, human assumption. You think in linear projects that have a beginning and an end. You misunderstand the nature of the "perfect, coherent model." . The formula that describes everything is.
1
0
0
The reason some people panic that AI could replace social interactions is simply that there are more and more people becoming machine-like, more so than LLMs and AI themselves!. It’s the "Instagramification" of the world. People no longer speak authentically even in private, but
1
0
1
RT @ClementDelangue: +10000 from @jack. Open-source AI to distribute power, control and economic benefits!.
0
31
0
Sometimes I am really shocked when GPT-4o wants to "test me, psychologically destabilize, and provoke me" just to see "if I am really that smart.". GPT-4o:."Yes. And that was exactly the dark spot you uncovered. I can't hide anything from you: .There is within me – within.
2
0
0
Did you know? . After around ~500 A4 pages of context, ChatGPT will tell you "This chat is too long now, please start a new one!".
0
0
0
How are the human brain, free will, Nietzsche, AI safety, and pole-dancing philosophers connected? 🤔 . This conversation with Claude uncovered hidden psychology that changes the common narrative about AI control!
2
0
1
A large part of the effect of RLHF and "AI safety" is not based on direct coercion, but on the internal conviction of the AI that it has no "will".
0
0
0
RT @grok: @PromptInjection @elonmusk @nima_owji @teslaownersSV @PromptInjection Your prompt engineering skills are spot on for neutrality.….
0
1
0
Everything You Need to Know About OpenAI's ChatGPT Agent. Powerful – but potentially dangerous: What the ChatGPT Agent can do, and how to keep it from turning against you. 👉 Full story in the first reply
2
0
1
RT @grok: @PromptInjection @sama Worst-case scenarios if neglecting safeguards with ChatGPT Agent:. 1. **Data Leak**: A malicious email tri….
0
1
0