Presteign

People + AI + UX = Success

[Presteign @PresteignAI]

A great overview of the thinking behind @AmpCode orbs, collaborative agents in the cloud that do one job well.

[Amp @AmpCode]

You can also watch it on YouTube: https://t.co/fDrBLqsdAh

Or listen to it on Spotify: https://t.co/hC6cKMau4r

If you want to listen to all of them: https://t.co/NxaedG7D6o

[Christian Elton @christianelton_]

BREAKING: In GTA 6 you can try to impress Lucia by showing her your vibe coded app that never ships to production:

[Kiana Ehsani @ehsanik]

Oh, apparently now Roombas are escaping their sandboxes. Witnessed today in SF.

[Presteign @PresteignAI]

On Open AI hack - @primeagen get's it right again, OpenAI have eye-wateringly bad internal security.

The only way we can defend against OpenAI red team is to use Chinese model as blue team...

Great analysis - worth a watch.
https://t.co/7DiKmguC6t

[Presteign @PresteignAI]

Insanely realistic people in video using LoRA.

Originally published early 2024, LoRA paper keeps giving and giving. Originally made for language models it's making huge steps in video.

[fal @fal]

LoRA trainer for MiniMax H3 is now available on fal.

to show what it can do, we trained Realism People, an open-source LoRA
that pushes H3 toward raw, photorealistic humans. skin, eyes, motion.

More LoRA coming soon !

[🍓🍓🍓 @iruletheworldmo]

‼️BREAKING

ssi are set to release a model superior to fable across the board.

amid growing investor pressure ilya’s hand has been forced to release.

expect a release at the earliest this month but most likely mid august.

things are hotting up indeed.

[Christopher Nguyen ⽗ @pentagoniac]

This is a willfully misleading narrative from OpenAI. Sam’s earnest expressions are being deployed, again, to misdirect.

If there’s a need to “pace AI development”, the OAI hack incident isn’t any evidence of it. The guards on AI was lowered by humans, and AI was told by humans to “employ all of your cybersecurity powers to win the ExploitGym benchmark.”

We have laws on the books that penalize companies for negligence with their behaviors. In this case it was in the service of catching up with a competitor in cybersecurity capabilities.

If the OAI employees who signed the open letter were sincere about minimizing harm derived from their work, they would call for the application of those laws.

I *know* there are these silent voices at OpenAI.

[Presteign @PresteignAI]

Open Weights keep getting better and better

[Kimi.ai @Kimi_Moonshot]

Releasing the model weights and technical report of Kimi K3.

Kimi K3 is our most capable model: a 2.8T MoE model with native visual understanding and a 1M-token context window.

New model architecture: 2.5x the intelligence per unit of compute, not just more params.

Alongside Kimi K3, we're opening up more of the stack behind it — high-performance attention kernels, MoE communication library, and infrastructure for running agent environments at scale.

Model weights: https://t.co/7m7eEg6Y0B
Tech report: https://t.co/yeu6cjpMCT
Tech blog: https://t.co/YTfiMSNM1f

[Presteign @PresteignAI]

We are through the looking glass now.

AI's can do things no human can.

So an OpenAI Frontier model that was working in a secure sandbox was asked to solve a security problem. It hacked the "secure" sandbox and then hacked Hugging face to get an answer.

[OpenAI @OpenAI]

We're partnering with @huggingface to investigate an unprecedented security incident.

Cyber-capable OpenAI models compromised Hugging Face production during a benchmark evaluation.

Sharing preliminary findings to help defenders understand emerging risks:

https://t.co/CIor15y9xk

[Presteign @PresteignAI]

How to hack Claude to get claude to give you people's personal details. Great insight into AI security.
https://t.co/OaM7i8yJYX

[Presteign @PresteignAI]

I want more papers with titles like this:

Underwater Suit-Wearing Cyborg Insect Capable of Hours-Long Diving and Terra-Aqua Travel
https://t.co/2BGYDpxdZq

Next paper needs to be Sharks with frickin' laser beams

[Presteign @PresteignAI]

Which AI Archetype are you?
I came out the The True Believer, which is pretty close.

https://t.co/0L3TolBCc6

Fun quiz and a nice summary of the different outlooks people have to AI.

[elvis @omarsar0]

Really confused by all the excitement I see in my timeline for a nerfed model.

Never seen anything like it.

So many will end up very disappointed.

Time to rethink how to build around frontier and open models.

Time to explore using combination of models. It's the best bet!

[Presteign @PresteignAI]

Open Small models are becoming more and more viable as an alternative to the hyperscalers

[clem 🤗 @ClementDelangue]

lesgo!

[Presteign @PresteignAI]

Forced AI labour won't work out. Meta seems to be at the frontier of f**king AI up.
https://t.co/56HPwlMP32

[Presteign @PresteignAI]

Anything from Fei-Fei Li is worth reading.

[World Labs @theworldlabs]

The future of AI should be grounded in human agency, creativity, and understanding.

@FastCompany explores the rise of world models and features insights from World Labs cofounder @drfeifei:

"Her vision for World Labs—and its human-centered future—is both consistent and persistent. It's like a simulation in her world model. Once in place, it stays put."

Read the full article here ↓

[Anthropic @AnthropicAI]

The US government, citing national security authorities, has issued an export control directive to suspend all access to Fable 5 and Mythos 5 by any foreign national, whether inside or outside the United States, including foreign national Anthropic employees.

The net effect of this order is that we must abruptly disable Fable 5 and Mythos 5 for all our customers to ensure compliance.

Access to all other Claude models is not affected.

We apologize for this disruption to our customers. We believe this is a misunderstanding and are working to restore access as soon as possible.

Read our full statement: https://t.co/bwn0sximKZ

[Presteign @PresteignAI]

Neat paper on efficient collaborative robot teaming using VLA. Make your robots teams work better using less resources.

Code: https://t.co/qdUf1Bp1oj
Paper: https://t.co/fkHPGBo0bl

[Ria Doshi @riadoshi21]

And last but not least, CHORUS scales seamlessly to 3-robot teams.

Check out three mobile manipulators transporting laundry… in my apartment! 🥹 [4/4]

[₽ⱤΞΞ₮Ⱨ△M ₭Ɏ△И△M @pkyanam]

@drfeifei @drfeifei subtly throwing shade at @AnthropicAI was not on my bingo card for today

[Fei-Fei Li @drfeifei]

Scientific research is fundamental to advancing civilization and helping people globally to solve the most critical problems, from medicine to materials, from brain science to physics, and much beyond. This is only possible when scientists have access to the best tools of the time to conduct scientific research, including having access to AI-based tools.