Gerry

Research Scientist In AI and Machine Learning, Posts only represents opinions of myself.

[Gerry @btc_gerry]

@yangxilin0483 @IamRamenPanda eth的估值本身就是严重高估的，得跌50%到合理估值再来谈拉盘的问题

[Aakash Gupta @aakashgupta]

Ilya said the quiet part out loud on Dwarkesh's pod, but most people still aren't processing what it means.

Here's what's actually happening inside AI labs.

Research teams have entire divisions that do nothing but create new RL training environments specifically designed to boost benchmark scores. They treat AIME, SWE-bench, and MMLU like standardized tests. The model practices 10,000 hours on competitive programming problems until every proof technique is at its fingertips.

Then it fails to fix a simple bug in production without introducing two new ones.

Sutskever used the perfect analogy. Student A grinds 10,000 hours of competitive programming. Memorizes every algorithm, every edge case, every proof technique. Becomes the #1 ranked competitive coder in the world. Student B practices 100 hours but has "it." Intuition. Taste. The ability to learn new things quickly.

Who has the better career? Student B. Current AI models are all Student A.

The benchmark gaming runs deeper than most realize. Studies have shown data contamination inflates model scores by 20-80% on popular benchmarks. The training-test boundary is porous. Models memorize answers rather than learn concepts. And when you control for contamination, much of what looks like intelligence is pattern-matching on seen data.

This explains the economic puzzle Ilya pointed to. Models score 100% on AIME 2025. They hit 70%+ on GDPval beating human professionals. Yet businesses still struggle to extract value. The benchmark performance says genius. The P&L says otherwise.

The sample efficiency gap tells you everything. A human teenager learns to drive any car after 10 hours. An AI model might need millions of examples and still fail on slight variations. A human learns a concept once and applies it everywhere. Models need to see the exact pattern thousands of times and still choke when the formatting changes slightly.

Sutskever's diagnosis: we're moving from the "age of scaling" (2020-2025) back to the "age of research." The belief that 100x more compute would transform everything is dying. His $3B company SSI is betting that the next breakthrough comes from solving generalization, not stacking more GPUs.

The labs know this.

That's why the benchmark arms race is accelerating.

It's easier to show impressive numbers than admit the fundamental approach might be plateauing.

[阿毛 @LmhdbAi]

坏了，是恋爱的感觉😨 @wolfyxbt 狼哥求你了

[Gerry @btc_gerry]

@rich_adul 目前除了赌博，洗钱之外看不到啥别的用途啊

[Gerry @btc_gerry]

@IamRamenPanda 你这纯属想当然了，只要是这美国运作的资产交易商都应该发税单，跟是不是证券没有啥关系。 只是这些平台根本不承认自己在美国运作，屏蔽了美国ip外加用户许可协议上不准美国用户。

[Gerry @btc_gerry]

@hantengri it is better than dot, ada, xrp, xlm , hbar though

[Gerry @btc_gerry]

@scipopieck @lidangzzz 还是比美国的本科生好不少的

[Gerry @btc_gerry]

@RZLHongKong @Summary52997362 @LiuZhongjing 你的这个说法仅限于对二元选择题成立。对于一个有无数种候选答案的问答题，0.9%的准确率说明他是真的不懂。这个在数学上叫做mutual information， 刘仲敬的答案跟正确答案之间的mutual information几乎是0.

[Gerry @btc_gerry]

@KylinSu3 @tj_research @WangNextDoor2 不懂就别吹牛逼了，论foundation model的实力，google甩了特斯拉八条街，是什么给了你质疑waymo技���路线的实力，真的笑死人了

[Gerry @btc_gerry]

@tj_research 就一个卖破衣服的公司，过去几年pe 50-60，现在算是价值回归

[Gerry @btc_gerry]

@mandate2049 硕帝在国内没被抓起来？

[Gerry @btc_gerry]

@pana067 @prjx_hl can they rug pull?

[Gerry @btc_gerry]

@DamiDefi web3 ai are scams

[Gerry @btc_gerry]

@lanhubiji 增速最快的，只有hyperliquid

[Gerry @btc_gerry]

@Balloon_Stocks = = 你们炒股票也能炒出这么大仇，又不是键政

[Gerry @btc_gerry]

@OraProtocol rug pull

[Gerry @btc_gerry]

@stevenyuntcap what's its pe now?

[Gerry @btc_gerry]

@zengjiajun_eth these marketplaces have no moat