What is the best AI ChatGPT or Copilot, Cursor, Grok?

Or will it remain an endless remix of the past data, experiences, observations, theories and models.

So the question is: will AI ever be capable of representing, reflecting and reinventing REALITY?

An analysis of eight AI models, including OpenAI's ChatGPT search and Google's Gemini, by researchers at the Tow Center for Digital Journalism found that “they gave an incorrect answer to more than 60% of queries.”

Hello, I have a question about astral projection. I started to get interested in this a little while after my mum passed in april. I thought I may be able to see her and speak with her if I managed to achieve astral projection. Since this interest, every time i sleep on my back I go into sleep paralysis. However, I cant progress into astral projection because it is very scary for me as I feel like I'm suffocating when this happens. I panic and force myself to wake up. This only ever happened about once a year before this. It sometimes lasts a long time. This has happened about 3 times per week since my mum died, as mentioned on a previous post. I no longer try to go into it anymore(due to the suffocating feeling), but it still happens. I read that sleep paralysis is the pathway to astral projection. Why has this started to happen so frequently since simply taking an interest in it? Is this connected to the afterlife? I am concerned about it as I now cannot seem to stop this happening. Could it be my mum trying to communicate? Im asking due to more knowledge around this in this group.

The “most accurate model to emerge from these tests, Perplexity from Perplexity AI, still answered 37% of its questions incorrectly. The village idiot award, meanwhile, goes to Elon Musk's chatbot Grok 3, which was wrong a staggering 94% of the time. Impressively bad.”

The real threat isn’t smarter machines — it’s a dumber humanity looking for optimization, automation, efficiency and productivity at the human costs.

Transformer models like GPT are nothing more than over-optimized probability statistical data algorithms and software models.

Earth’s Seven Continents Might Be a Myth: Only Six Continents May Truly Exist After All - The Daily Galaxy

They don’t understand content and context — they compute and calculate, interpolate and extrapolate it. A large language model (LLM) takes your prompt, smart or not, draws from trillions of tokens in its training corpus, and generates the most probable responses.

Will it ever do something that shakes us to the very foundation, shifts our world, forces us to think completely differently?