The hitch is that the #LLM, because of the statistical and stochastic (i.e. randomized) methods by which it assembles te...

The hitch is that the #LLM, because of the statistical and stochastic (i.e. randomized) methods by which it assembles text, isn't actually capable of significant learning. The best the #GenAI sludge-merchants can manage is to try overpowering the innate tendency of the LLM to emit gibberish by constantly pouring more high-quality human prose into their training data, attempting to get a better emulation of human writing and speech through brute force.(cont'd)