Comment by R1D3R | Hacker News Reader

Comment by R1D3R | Hacker News Reader

Parent Full thread

I don't think you have adequately steel-manned the current anti-AI sentiments. I encourage you to consider more deeply why people would be opposed to AI as it exists in our current regulatory, political, and social climate.

As a starting point:

Business leadership is giddy to replace workers with AI.

Investors are eager to roll out massive resource-intensive and locally polluting industrial installations.

Politicians are eager to rubber-stamp these developments without community input, bypassing existing regulation and failing to satisfy or even address community concerns.

The defense industry is recklessly employing this emerging technology to make life and death decisions, already resulting in the deaths of innocents.

I also think you should fundamentally reconsider your comparison. Data is a singular being, not a widespread new technology. The only similarity between Data and 2026 LLMs is that they are not biological.

It is a fair point to say I wasn't steel-manning the best anti-AI arguments, but I think that was mostly because I wasn't actually trying to argue against ALL anti-AI arguments. I think there are definitely some thought provoking and somewhat persuasive anti-AI arguments, but I don't think those are the ones I was questioning.

I was speaking more of the general "I hate AI because it has no soul" types of arguments. I understand that is separate from the arguments you lay out, which is more a question of the current structure and implementation, rather than an outright anti-AI in general case.

"Soul" is one of those things that we cannot point to a specific factor or set of factors, but we know it exists. LLMs trained on human-produced content perform well, but LLMs trained on LLM-produced content undergo model collapse fairly rapidly. "Soul" is the quality that allows us to produce such robust training data. We each have a unique point of view derived from our life's experience and how that altered our brain, that comes through in whatever work we produce—not just a convolution over a large number of thin vertical slices of human points of view. Now, what this may mean is that our brains are just token predictors that constantly ingest large amounts of data from the real world and continuously retrain on that data (and do so on ~20 W of power which is phenomenal by current AI standards). In fact you can induce a sort of model collapse in a human by isolating them in a sensory-deprivation chamber, after which they will literally begin to hallucinate. So the creation of a machine with a soul is not theoretically off the table, but as it is—yeah, LLMs don't have it.

If you aren't arguing the best positions then I'm bored. There's an infinite number of bad arguments.

Reply on news.ycombinator.com