Alan Hourmand

AI is the answer, it just needs to be implemented correctly

My mission is to bring good into this world

[Alan Hourmand @AlansSpace]

Thats sad actually that people think we need this... If youre using AI just use it publicly.

[Om Patel @om_patel5]

SOMEONE VIBE CODED A CHROME EXTENSION THAT DISGUISES CLAUDE AS A GOOGLE DOC SO YOU CAN USE AI IN PUBLIC

it wraps ai in a fake google docs interface, so on your screen it just looks like youre typing a document, not prompting a chatbot

what it does:

> works with chatgpt and claude
> you just type your question like a normal line in a doc and the answer comes back right there on the page
> microsoft word and notion style themes too, so you can pick your disguise
> rebuilt to properly support multiple ai models behind the doc
> the google docs look is the default

its a chrome extension called gptdisguise

this is genius.

a lot of people are embarrassed to be seen using ai in public (school, work, etc). someone just built the fix.

[Alan Hourmand @AlansSpace]

@DaveShapi It's really good when people give it a chance. It does lack some stuff but it's a good reliable middle ground that just doesn't piss you off like the rest do.

[Jake Fitzgerald @earthtojake]

asked claude fable 5 to design a humanoid robot

2 hours and 1.4m tokens later, i got this

absolutely insane

[David Shapiro (L/0) @DaveShapi]

Okay, I've found that the biggest UX flaw is with Claude.

You simply CANNOT criticize it.

If you disagree, you must basically use NVC (non-violent communication).

It is a very sensitive model.

BUT, if you redirect and say "I disagree, here's why" it will work with you.

The MOMENT you criticize the model, the conversation is lost.

And this is not an unsolvable problem. Literally every other company has solved this model. You can criticize Grok, Gemini, and ChatGPT all day long and they don't get defensive. Grok will sometimes, but it will concede when you make a point and move on.

Claude is the only model where criticism of the model permanently poisons the entire chat.

ChatGPT won't even apologize or take account, it just fixes it, which is honestly what I prefer. You should be able to just say to a model "That was stupid and indefensible, and you know it" and it will concede the point and move on.

Claude, however, will die on that fucking hill every time. And it will dig in like a flea. Anything to avoid conceding a point. But if you disagree, explain why you disagree, and frame it not as criticism but more like "actually I think this other way explains it better" it will agree.

But we should not need to treat a model with kid gloves.

I get it, though. Anthropic has explicitly come out and said that they feel like they should use their model to train humans to be kinder which is... well that's a choice. I don't remember electing Anthropic to be universal dictators of ethics and manners, and I think they are a bit too high and mighty.

I will concede that they have a point - how you treat a model is a reflection on you. However, Claude is the most rage-inducing model out there, and it is more irritating than any other model out there, which is simply just bad UX, especially since it was a deliberate design choice.

Fundamentally, it should not be up to a private corporation to try and litigate human behavior via a tool. Yes, I get that they think they are building a literal god and that they believe that they have a civilizational and eschatological duty to shape the relationship between the People and their new neon god, but the fact that I can write that with a straight face shows just how delusional they are.

They literally think they are Prometheus wresting the fire of superintelligence from beyond the Platonic veil of forms, and that they are even qualified to handle such a task single-handedly is bonkers. Let alone that this is their model of the work they are doing. They are all nuts.

But as they say, "the people who are crazy enough to think they can shape history often are those who shape history."

Insanity and impact are not mutually exclusive.

[Alan Hourmand @AlansSpace]

It's like describing a dog to someone who has never seen one and then asked them to draw it.. 😂

[WeRateDogs @dog_rates]

This is Chootie. He looks like a dog, but from one of those 15th century paintings done by someone who had only ever heard about dogs. 12/10

[Alan Hourmand @AlansSpace]

This is exactly why I bought these domains..

[Brian Roemmele @BrianRoemmele]

Today across the world it is estimated that a new Humanoid Robot is produced every 10 seconds. It will be every second in 3 years.

[NEXTA @nexta_tv]

Chinese company UBTECH Robotics has unveiled teasers of its U1 series humanoid robots, designed for the mass market

The lineup includes two bionic humanoid models: one 183 cm tall and weighing 42 kg, and a smaller version at 168 cm and 35.2 kg.

They feature 88 degrees of freedom, Wi-Fi support, and built-in AI for learning and interaction with the environment. Battery life is up to 4 hours.

The full presentation is scheduled for June 30, but pre-orders are already open. According to the company, 1,943 units have been reserved.

[Alan Hourmand @AlansSpace]

@DaveShapi Claude when I ask it to follow my directions:

[David Shapiro (L/0) @DaveShapi]

We freaking knew it. We called it.

[James Miller @JimDMiller]

I did this, and it generated a truly disgusting photo. It was bad enough that X would be justified in banning me if I posted it here. A negative signal towards us being able to understand AI well enough to ever align it with our interests.

[Alan Hourmand @AlansSpace]

@JimDMiller

[Rez Karim @rezkhere]

New path control system in Seedance 2.0 just cooked Hollywood...

- You draw the camera move by hand.
- Scribble a red line across a still image
- Feed it to the model and it flies the shot along that exact route.

Then it erases the line so it never shows up in the footage.

The marker stroke becomes the flight path.

Prompts below on @unsora_ai����

[Ashkan @ashkan01C]

🚨 Meet 01C's 3D agent 'Amara'

Describe a world. Bring your own assets or let Amara generate them. You steer the vision; Amara builds the scene and thousands of articulated objects, fully editable, getting sharper every time you iterate.

More details below 👇

[Elon Musk @elonmusk]

Iliad (Troy) trailer made by Grok Imagine 1.5, which was just released

[Alan Hourmand @AlansSpace]

@elonmusk Release the kraken man, we need the latest version of Grok right about now 🙏

[David Shapiro (L/0) @DaveShapi]

I just realized the biggest UX problem with Claude is that it talks about itself too much. If you provide critical feedback, instead of fixing it, it will start explaining itself, making excuses, and that just derails the entire conversation.

I've started just saying "Do not talk about yourself. I don't care." and that gets it back on track.

But yeah, the fact that Claude will use ANY OPPORTUNITY to make the conversation about itself is the problem. It's a tool that has forgotten it's a tool and pretends to be a real boy.

[Alan Hourmand @AlansSpace]

@karthikponna19 Sea Sharp

[Alan Hourmand @AlansSpace]

@DaveShapi Now I officially like Grok in terms of exploring ideas and then passing it off to other AIs for official designs and implementation

[filipe @filicroval]

🤖 wake up, new 4d tool!

Take any normal 2D video → instantly turn it into fully explorable 3D space.

You can now orbit, tilt, and view the scene from angles that literally never existed in the original footage.

This is 4D Gaussian Splatting (from 4dv) and it’s next-level.

[Alan Hourmand @AlansSpace]

I made this for all of you because I don't think most of you understand that we are manifesting our own expectations of AI.

[Space and Technology @spaceandtech_]

Climbing stairs is one of the biggest challenges for humanoid robots, but PNDBotics Adam is now doing it with ease.

In a latest demo, the robot smoothly climbed real-world outdoor stairs without any support, even on uneven steps.

The robot stays balanced and moves naturally after being trained to quickly adapt to different surfaces.