Jill Nephew

The VP of a private investment firm, Gloria Caulfield, speaking at UCF’s graduation: “The rise of artificial intelligence is the next industrial revolution.” The entire graduating class responded with loud boos highlighting the growing frustration with AI rapidly replacing jobs. The U.S. ruling class is so insulated from reality they cannot see the revolution that is brewing for a new way of life that prioritizes our people and our communities over profits.

@GaryMarcus @geoffreyhinton What I really don't understand is how this is getting so complicated. A basic understanding of statistical models includes the concept of over-fitting if the parameter space is larger than the data space. Of course this will happen when data is uneven, how could it not?

Am old enough to remember when @GeoffreyHinton told me I was stupid for saying that LLMs regurgitate training data. He was wrong. LLM regurgitation is now one of the best-established findings in the field. Excerpt below from a new DeepMind paper; every single one of the papers shows that Hinton was wrong. (Also: still waiting for AI to replace radiologists.)

Banjo folk musician: "They fed YouTube videos of me to an AI engine, that then mimicked my voice and playing... Vydia has come forward and made copyright claims on my YouTube page... I'm no longer making money on YouTube. Vydia is...over AI-generated music."

@MrEwanMorrison @null_ropex Interpolate is a non-anthro term. And it describes what curve fitting is.

@null_ropex ah, but that would be more anthropomorphism

I'm not going to use the word "hallucinate' any more to describe LLM AI. It was a trick to use this word from the start. First, the AI pushers deliberately used an anthropomorphic word to describe machine error. Then when you point out the machine errors, they say, "hey, humans hallucinate too." Then they even have the audacity to say "well, if this AI hallucinates it's proof that it must be sentient!" The acceptance of this trick-word has been an error from the get go. Offer me some alternatives words, please. Accurate and derogatory words are both accepted.

Just think, if every investor had gotten a proper education in statistics, there would be no AI bubble. Education matters.

@DoctorPerin @New__Ruth So maybe you are saying these people aren't autistic, but there would be no way to probe if this choice is made privately and there is no trust in the researcher to reveal that because they are part of the problem.

@DoctorPerin @New__Ruth I should clarify - "social relating" as a means to gain social equity/promote social cohesion. Generally group goals/behaviors are ethically problematic in ways "outsiders" can see, so there is that too.

“High-masking” requires a sophisticated understanding of higher-order social skills to calculate effective superficial responses. The more skillful someone is at this, the less likely autism is the most accurate diagnosis.

@bnielson01 @leecronin @perigean @michaelpollan @grok Huh. No snarky comeback this time. Imagine that.

@bnielson01 @leecronin @perigean @michaelpollan @grok Or this shorthand:

"Unquestioned ... is the metaphor that brains are computers—the hardware on which the software of consciousness is run. Here, we meet a metaphor parading as fact. Indeed, the whole paper and its conclusions hinge on the validity of this metaphor."

@bnielson01 @leecronin @perigean @michaelpollan @grok People use the term "word salad" when they encounter someone that is more well read on topics they are and can't believe it. My last challenge to you is sufficient to disprove the brain is a computer - prove the brain isn't computed. The rest will follow.

@jillnephew @leecronin @perigean @michaelpollan @grok I'm not really surprised that you took this position.

@bnielson01 @leecronin @perigean @michaelpollan @grok I cannot make these arguments going back decades for you on twitter if you aren't familiar with them. Sorry.

@bnielson01 @leecronin @perigean @michaelpollan @grok I assumed you had encountered these arguments before and I could shorthand. It appears you haven't. Here's another one: how can you prove the brain computes versus the brain is computed by the environment?

@bnielson01 @leecronin @perigean @michaelpollan @grok So the brain is not a computer, the brain is an emulator that can emulate a computer (some can).

@bnielson01 @leecronin @perigean @michaelpollan @grok To escape triviality, you "close" the system (Markov blanket etc etc). I argue above, yes, this is the context where brains can emulate (and I mean emulate) a computer. But that is not what they are, it is what they can do, but they cannot do it and live.