Hacker News new | past | comments | ask | show | jobs | submit | _giorgio_'s comments login

He can't even press the shift keycap anymore.

Clearly you don't use models so much.

Even in the openAI ecosystem there are models that, while similar in theory, produce very different results, so much that some murderous are unusable. So even small differences translate to enormous differences.


I use AI everyday for work, mostly models from OpenAI, Anthropic and DeepSeek. In my experience none of them completely dominate the others. You seem to disagree strongly but then just state your argument, which model or company do you think is the clear leader currently and why?

The AI race is super close and interesting at the moment in my opinion.


O1

People using AI, even the high school teachers that I know, constantly compare and battle models against each other. Even a 10% difference in the results is something that it's worth paying for, because it saves you a lot of time

Tesla makes backup generators.

Do they? Powerwall stores power, but doesn't generate it.

You need to offer AI just to stay in the business, whatever is your business.

Just look at how much money Google lost in that failed AI demo from 2003.

The stock would be worth 50% less if the invested nothing in AI. Even the founders are back because of it.


https://x.com/svpino/status/1592140348905517056

""" In 2017, a team led by Andrew Ng published a paper showing off a Deep Learning model to detect pneumonia.

[...]

But there was a big problem with their results:

[...]

A random split would have sent images from the same patient to the train and validation sets.

This creates a leaky validation strategy.

"""

He's not infallible.


Deepseek spent at least 1.5 billion on hardware.


Grok is the best LLM on https://lmarena.ai/.

---

No benchmarks involved, just user preference.

Rank* (UB)Rank (StyleCtrl)ModelArena Score95% CIVotesOrganizationLicense 1 1

chocolate (Early Grok-3) 1402 +7/-6 7829 xAI Proprietary 2 4

Gemini-2.0-Flash-Thinking-Exp-01-21 1385 +5/-5 13336 Google Proprietary 2 2

Gemini-2.0-Pro-Exp-02-05 1379 +5/-6 11197 GoogleProprietary


Just an ignorant and non technical guy trying to stay relevant.

He has an infinite quantity of time and somehow always finds someone to interview him about the fact that AIs don't work and have been hitting a wall since the last ten years. He invents some metrics that nobody has agreed on, and says that they're below that.


You received a tool. A great tool, a magnificent tool.

Learn to understand its limitations and make the best use of it. Surely it's confused by lesser known facts, that's a thing that you can't ignore even if you interpret AI as a tool that compresses knowledge.

If you don't understand that, you're the tool.


The more salient point is that you might know the limitations of the tool, I might know the limitations of the tool, but millions of people who don't are using it for things it has known limitations for, because the marketing blitz that sits atop this glosses over those limitations.


The problem is that nobody reads the tool manual


Moreover, the manual also lies about the tools limitations.


Join us for AI Startup School this June 16-17 in San Francisco!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: