@feardustcr@raf_the_king_ I doubt they use it to continue to train the model in a way that matters. They are simulating thousands of games a minute. A few hundred real player games a day make virtually no difference. The ai is already better then ~95% of players so it can just train against itself.
True, you can’t stop a client reading its own RAM. But you can stop sending the data early. CR schedules plays ahead to hide lag, so the cheat sees them first. Fix: send each play only when it’s due, or late with rollback. The 1s deploy timer hides the delay, so there’s nothing to read early.
Or just make a detection tool, could be made in 30 min : vary delivery timing. A human reacts to when a card appears, a cheat to when the packet arrives.
We raised $5.3M to give AI agents a computer to run on, with their own email, phone, iMessage, and WhatsApp
Led by @peakxvpartners, with @rauchg, @sternhenri, @jefrankle, @cjc, and @daniel_levine
Today we’re launching hosted agents on @inkbox_ai. Your agent runs next to its communication channels, so it never misses a reply. It sleeps when idle, so you don’t pay for downtime.
In the 4 months since our YC launch, we’ve processed millions of messages for AI agents.
@Mike_from_PA I mean just from a buisnsss perspective no one is gonna watch the podcast if Ethan isn’t around l. It’ll just be burning money. seems like a reasonably smart business decision.
I can promise you after reading our first two blatantly ai generated slop answers they are gonna move on. These people get thousands of applications, You’re not gonna stick out by filling the response with air.
I highly recommend you review previously accepted applications and try and write most of the application yourself.
🚨 Nuisance streamer Onyxdenji tried to give away fried chicken in the hood and FOUND OUT after calling a black man the 'N WORD' 💀😐
ChudTheBuilder copycats are now popping up across the United States at an alarming rate...
I feel like sooner than people think, every major company is just going to expose a direct layer for AI to act through.
At that point, years of work and millions invested into custom agent harnesses could become something GPT or Claude can do almost overnight through direct connections.
I genuinely don’t understand the massive investments into this layer.