drivebyhooting

“Note there is also a difference between knowing mathematics deeply and explaining it fluently; they are related but different skills. Ideally I want both, but it is the first that I ultimately value more highly.”

Once again perpetuating the cardinal sin of lionizing mathematics heroes. Which to be fair is just THE culture of higher mathematics.

What good is a minuscule elite Cabal of savants? If these deep insights are inaccessible to 99.999% of the population, then they are no better than the manuscripts written by AI. We already have those.

Fresh sessions don’t help because recent refusal context is injected across conversations in a clandestine way. I figured this out myself through reverse engineering these POSs.

I feel like the solution presented is just a clustered hashmap or “locality sensitive” if you’d like it.

Still not orders of magnitude better.

I hope so. I would never bow down to such a god.

One way to look at the current AI economics is that consumers and businesses are subsidizing training for super intelligence only available to government.

Do you think Claude and GPT refuse doing calculations for the national ignition facility? Or for STRATCOM?

Mark my words: AI will only be allowed to destroy the software engineering profession, but somehow protectionism will be enforced for everything else in the name of “safety” and “fairness”.

Antigravity harness seems much worse than Codex. Argon seems very knowledgeable but very sloppy. If I force it to self review it will catch its own errors and laziness.

Don’t be daft. All that man power would be excellent for bolstering the army and invading neighboring countries.

I imagine a human body will still be cheaper and more agile than robots for a long while.

How can anyone with children be intense (at work) is beyond me.

Just talking about salaries, revenue, and stock IN AMERICA.

Doubtlessly Chinese manufacturing benefited tremendously from those advancements. But my point is the headline whinging about AI in America would also apply to Tech in America.

End of ZIRP, tax law changes, post-COVID inflation, and general economic malaise from federal mismanagement.

There are too many confounders.

I think you could say the same thing of tech and the economy prior to AI.

No Signs of Tech Payoff Outside the Tech Sector. Manufacturing, basic science, education, etc languished while Big Tech reaped record profits and achieved stock heights for the last 15 years.

Article seems to answer the question: because generated text has the competing objective of information density and supporting reasoning chains.

But I also wonder why AI loves to add slogans, catch phrases, and witty sentence fragments. It’s grating and a dead giveaway of slop.

Good thing IQ has nothing to do with solving novel programming or architectural puzzles :)

Verbalizing my thoughts as I try to solve a problem drops my IQ down by like 20 points at least.

Cool when asked what I do I’ll say I’m an unemployed code monkey instead.

You’re just saying that because you think you’re clever, making your little mathematical quip rather than looking into the actual data and argument.

People clearly have preconceived conclusions and they don’t have an interest in learning any facts.

Look, I’m not going to rebut each of your points one by one. Ask AI to steelman both sides and learn the facts for yourself. Then you can decide.

But that last point about SAT is just too much for me. Just because the study controlled for SAT score doesn’t mean private education does not impact SAT score. In fact there is a correlation. And even causation. You can dig into the data yourself.

For sake of conciseness I didn’t explicitly say elite non-religious private school.

Please check your rebuttals against AI before you post it.

I don’t know why humans even bother talking to each other anymore. Go ask AI why your argument is wrong.

I already did it for you: the numbers I gave you are already conditioned on all that. You’re making the false assumption that moving from each tier is a low probability event. It is not. The retention from elite elementary to elite high school is very high. It’s not some hyper improbable funnel.

No. Spar this with AI. The 2X multiplier is AFTER controlling income.

What you’re saying is just a comfortable lie to make people feel better about public schooling.

False. It massively increases access to good teachers and ultimately university.

Of course you can always go to a no name CSU. I’ve been on both sides of the divide. The education at a cal state is like elite high school or lower.

You can ask AI. Access to elite colleges is super conditioned on going to private elementary schools. North of 2x.

To simplify: C = elite colleges

P(C| private high school) = 2 P(C| affluent public high school)

P(C | private elementary) = 5P(C | affluent public schools) = 35 P(C)

Yeah the public school that serves reheated fast food and bowls of cereal for lunch. I’m not making it up. That is the district wide menu.

All well and good until they’re 5 or 6 and you find out all school admissions are gated by tests and performance. Oh unless you’re old money. Then your kids get in.

First of all, what happened to Max and Ultra?

Secondly, I’m mostly using Astra Ultra or Sol 6.1 when tight on quota. I pay $200/month and rarely if ever run out. Am I doing something wrong? Or perhaps I’m not that productive as you guys?

LLMs are amazing for engineering work. They can automate design and testing of the entire spectrum.

One aspect that currently frustrates me is that even paying for the highest plans and spending multiple quotas across all flagship models, I still haven’t been able to reproduce any of the amazing feats Anthropic and OpenAI gush about.

I suspect their internal harness, models, and thinking budget far outstrip anything us peasants have access to (never mind the compute). So saying that they achieve all these headline results with 3 hours of GPT Pro is disingenuous. If that were so, hundreds of math phd students would’ve already recognized it.

I’m coming to the realization that my monthly subscription is only helping subsidize their real internal AGI, and my interactions are a small amount of human feedback and creativity which they can strip mine should they choose to.

In terms of safety and alignment, how is this different from looping or increasing the depth?

Astra still outperforms Argon for me. And with a lot more quota too.

I hope Google can catch up to OpenAI. The more competition the better.