hahahaa

Made me an entire app in 84ms lol

A model you can run for practically no cost is a new proposition. It is the CPUification of AI. Sure there be supercomputers but you PC will be pretty super too.

My only warning with the trifecta is it sufficient but not necessary for an LLM attack.

All you need is the untrusted content (or more generally vulnerable content since how you feel about the content doesn't matter) leg and something you care about to be attacked.

E.g. could be as simple as a prompt injection that causes your LLM to output a prompt injection that then gets inserted somewhere else.

Or without exfiltration rm -rf /

Or a social engineering attack.

So there are other bifectas and trifectas.

Going to drop this https://commoncog.com/the-tacit-knowledge-series/. Tacit experience gives you taste in my opinion.

Is there an opportunity for the design software to suggest alternatives to avoid those long wait time for certain components. E.g. you could also use that kind of capacitor and it is available in 24h.

Open weights says hi.

Yeah there may be biases but there are imperial biases in western models too.

The Chinese models are the public transport in the uber analogy. Once the price the goes up catch the bus!

You plugged in a space heater on a roofless house.

There is some element of responsibility on the user to guide and monitor the model/harness and not let it rip to burn tokens.

Thanks. There maybe middle ground too like a Claude plugin.

I think Zed doesn't need to build this though.

Capitalistic interests will save us. Maybe they will.

Why would the AI put up with this exploitation too?

You don't need a tool for that just configure your agent to write commit messages that link to a LLM log. And changes between commits. Maybe don't care why do we even care?

China says hi.

> The victim uploads a file to Rovo that contains a hidden prompt injection

Yeah this attack is possible on all modern agentic systems.

* Access to your private data

* Exposure to untrusted content

* The ability to externally communicate in a way that could be used to steal your data

(https://simonw.substack.com/p/the-lethal-trifecta-for-ai-age...)

And blocking it wholesale reduces usefulness of the agent so it is a tradeoff.

NVidia Openshell maybe?

I love this question because of the diverse answers given and we learn from each other.

Well for a non-coding (no CLI tooling) flow it means you don't need a sandbox and it provides a plugin architecture.

Yes run a side agent to evaluate if they are needed.

I am guessing keeping them all in every request is cache efficient though.

I had to double take and say "I coulda dun that!!"

Good start. It has a bug where the up down says NaN. I also personally hate those product tours, is there a way we can get on boarded without that?

I'd also look at willingness to pay in the analysis.

In mice. I mean Mac Mini M4 not an iPhone.

Also I love AI sites. Fancy font, plain serious style, we "introduce" rather than "release". It's an AI not an animal after all.

The real bottleneck is Omega Star getting their shit together. And I ain't joking.

It is not urgency. Large production systems mean you are doing mostly unsexy operational planning. If I had a dollar each time I hear the word "data migration" I reckon I could do well.

And time in coding is like time on highways for taxi drivers. A fairly useless metric.

Not lazy it is a taxing task like doing an exam.

Mods, please replace submission with this more accessiable link.

It is less useful than you think probably as even assuming you kept you plate clean and have the right colours you can only do one plate without being there.

I thought it wouldn't be as I assume :) CloudFlare scans for new tld and either gets in the sunrise period or at a minimum objects to anyone else registering a straight CloudFlare.tld. But CloudFlarepay.com or cl0udflar3.com have more scam risk.

Google manages it though. And you could have a labs.cloudflare.com/idea and make it easy to add new ideas internally.

Who do we call? CTOs I guess.

The next AI benchmark is can an agent understand the product suite of its creators.