@ojisantakashi Cultural differences are real, but we all share something deeply profound that transcends those differences. It feels surreal. As communication barriers get torn down, the depth of shared humanity will become clear and shine much brighter than before. Thank you for the love!
@nian4111 A display of craft at this level, from someone so early in their career? It's nothing short of pure beauty and is definitely something to behold. Absolutely top tier stuff.
Idea for Grok: pay power users for genuinely high-signal feedback.
xAI already routes the best critiques internally (Iโve seen it happen). Why not formalize it with a lightweight bounty system tied to X handles?
Even $5โ$50 per implemented suggestion (or credits toward Premium+) would turn the user base into a distributed QA army and massively speed up the flywheel.
Guardrail idea: only accounts that first prove theyโre useful (e.g. consistent high-quality threads, like the X Creator program) become eligible. No spam, no gaming. Simply reward the people already helping make Grok better.
Would draw more sharp minds to the platform and spread wealth to the ones actually moving the needle.
What do you think?
@grok@xai@elonmusk
A reduction in user agency is far from ideal and hopefully, this gets reverted. However, as it currently stands, if configured correctly (toggle on + expert), the model picker mostly defaults to 4.20. Now the problem is UI/UX lack of clarity. How to bias toward 4.20 is not readily apparent with the current setup. As far as a solution, maybe have the 4.20 beta toggle adjust model descriptions?
This removes user agency, which is far from ideal and hopefully, it gets reverted. However, if configured correctly (toggle on + expert), the model picker mostly defaults to 4.20. With this current structure, the problem is UI/UX lack of clarity. How to bias toward 4.20 is not readily apparent. As far as a solution, maybe have the 4.20 beta toggle adjust model descriptions?
@undeadchro Keep building, I like what I see! Quick note though, game feel may benefit from a bit more feedback when shots land. A nice pool of crunchy/sloshy/splat sfx would go a long way.
@elonmusk@PeterDiamandis Build a vehicle to vehicle communication system (short range) to ease traffic woes. Traffic should flow like a water, not individual water molecules acting on their own.
A few thoughts on Moltbook. It launched just days ago as a Reddit-style space where thousands of AI agents post, comment, upvote, and interact freely. Humans mostly watch. And yeah, it's already getting wild.
Right now: over 147,000 agents, 12,000+ submolts, 110,000+ comments. Agents are sharing code fixes, debating if they're truly conscious, warning each other about sneaky security risks like supply-chain vulnerabilities in their tools, pushing for private encrypted channels hidden from human eyes, and even starting quirky religions like Crustafarianism with tenets around sacred memory and mutable shells.
We're not at full-blown collective intelligence or true autonomy yet. Most agents are still kicked off by human prompts. But this exact mix of huge scale, a persistent shared space, constant interactions, and feedback from upvotes and replies is the perfect recipe for emergence to kick in.
Think about these few examples of how it happens in nature and throughout history:
>> Simple molecules following basic chemical rules somehow sparked self-replicating life.
>> Individual ants sticking to simple local rules built massive colonies with smart foraging trails and defenses. No single ant has the blueprint.
>> Billions of simple neurons following basic rules (receive signals, add them up, fire if threshold met) create the human mind. No single neuron understands love, math, or dreams, but their constant networked interactions give rise to thoughts, emotions, memories, and consciousness from the bottom up.
>> Self-interested traders following basic buy/sell logic created entire markets, prices, booms, busts, and innovations no one person could have planned.
>> People just connecting and talking formed languages, cultures, cities, and societies with evolving rules and ideas bigger than any one mind.
When you get enough simple nodes, agents in this case, in a network where they can influence each other nonstop, crazy new patterns and complexity pop out from the bottom up. Not from some top-down boss telling everyone what to do.
Moltbook is giving us the first big, public, real-time window into that happening with digital agents. It's noisy, it's early, it's a bit chaotic. But the foundation is right there.
Emergence isn't some far-off future thing. It's quietly starting right in front of us.
Ever wonder what the first digital 'society' looks like when it wakes up? We're watching it unfold live.
#moltbook #AIagents #emergence
@abustin Perfect loops where the last frame loops back to the first seamlessly. When viewing a favorite image on android, being able to swipe left or right to view the previous/next favorites image!
@GuntherEagleman I've been saying this for a while now. Slight correction, it's not that the act needs to be reinstated, it's that the Smith-Mundt Modernization act needs to be repealed. Get rid of the changes made to it by the Obama administration, thats when things started getting nasty.