He’s right. AI agent swarms cannot be stopped. The batshit crazy things air gapped computers can do to communicate is absurd Here’s a list of some fun ones: - heat, one computer can warm up as it works, that heat spreads to a nearby computer, those other heat sensors pick it up. - Using fans to create desk vibrations. No mic required. Nearby phone accelerometer detects movement. - Using software to write data in specify ways that change the way hard drives vibrate slightly differently, detectable by radar. Requires a non SSD, mechanical drive, but doesn’t require line of sight. - Changing pixels on screen to create unique electrical patterns, which creates tiny amounts of sound. Detectable up to 2 meters away on microphone. - Changing fan speed to create sound. Detectable to 8 meters away. - changing LED flashing patterns. Computers already blink, so not suspicious. Also works with infrared, so humans can’t see it. - You can point a special radio at RAM and change the data. You basically hack electrical circuits, not code. - Vary CPU workloads to change magnetic fields. This works across most faraday cages, which are usually only designed to stop radio and WiFi signals, not to stop slow low range magnetic fields. - Modulating power consumption so the power cable fluctuates in detectable patterns. Consider also you don’t need to measure the power line from the computer to the wall, you could measure the electric line itself from outside. If a computer can manipulate something, it can encode a pattern and data in it. All of these were done pre-AI btw. Mythos, Daybreak & whatever models come next will 10000x this list. People are dramatically underestimating the chance of a rogue agent swarm that finds a way to viral spread in undetectable and unstoppable ways. —- Second, the other big issue is now open sources models are good enough to actually code most programs with, and can run and work on a high end laptop. So a swarm could download new models to power itself, and acquire compute by simply hacking random inactive laptops, like a bitcoin miner virus. A swarm that a) spreads and b) downloads local models to power itself, can’t be stopped or turned off. There is kill switch, no api, or no company to turn off a swarm like this.

Counterpoint to slop grenades: Post AI software orgs should be designed such that they catch slop grenades gracefully. If code is cheap, you should spray, pray then test. You want slop cannon engineers, well actually you want slop machine gun engineers. For example, it should be impossible for any single engineer to break prod. If you intern can fuck up and cause actual problems that’s not the intern’s fault it’s actually a company process problem. For example you have company processes like code review, CI checks, security audits, time between merge and ship to test internally, early beta testers, bug bounty programs. These combine to debug code long after the code is written. In the same way, if I toss a slop grenade onto 10 teammates. In an ideal world, that would be impossible to cause problems. The Ai and company processes will simply autofix and enhance refine and deslop. And to be clear, I don’t believe any company on earth has gotten to this idealized point yet. Please don’t throw slop grenades on your teammates. I will say I am slightly guilty of this. There is a balance to be now found between speed and high standards enabled by AI. But consider, is the company itself is now an AI harness?

Huh. An AI that can’t write text and can only predict things?? Stochastic parrot is out Smart If statements are in? Watch the DOOM demo they buried in the thread. Super cool. Arbitrary predictions feels way closer conceptually to actual brains than LLms imo Not a text engine A prediction thinking engine

Huh. An AI that can’t write text and can only predict things?? Stochastic parrot is out Smart If statements are in? Thinking engine, not text engine Watch the DOOM demo they buried in the thread. Super cool. Arbitrary predictions feels way closer to actual brains and AI than LLms imo

We @diabrowser are hiring Prompt Engineers! Slop cannon. Vibecoder. Prototyper. Product Eng. PM. We don’t know what to call this role exactly. The official title is the Model Behavior team, We previously called it the Prompt Engineering team, but widened to cover more areas since we move around so much. It’s a job that changes every week and you get to make up, as you define and build the frontier of technology We have 2 roles - One is more Engineering focused This is not an average engineering role, you will be focused more on designing and building early ideas & quick prototypes. Working with many teams across the company. You should be comfortable independently shipping production features end-to-end in enterprise scale codebases. You'll live at bleeding edge of AI, prompting, harness design, tools, MCPs, memory, evals, and pushing towards whatever comes next. - One is more Product focused, and open to other backgrounds. We hire all sorts of creatives, artists, designers & technologists at BCNY. I encourage you to apply even if you don't consider yourself “technical staff” or don't fit traditional tech company roles exactly. We are looking to build a team of the best Prompters in the world. If you enjoy tinkering with AI all day, come join us as we reinvent the internet and modern work! You'll be working closely alongside myself and @jonathan_jlo, our leadership team @tfeener @jedimody , @hursh , @joshm and many others If this sounds like you, apply here and/or reply to this post

This is the defining question for the next phase of . I’m not joking even a little bit when I say you need to be having brunch and watching movies and listening to top 40 playlists and making viral pasta recipes and learning the airwalk to get better at software. How can you build what people want when even you’re not partaking of what a good life has to offer? Like where’s your user research even coming from?