kfish

March 29th in Puerto Rico @pizzaninjas are hosting the pre-party for Tokenize conference at our clubhouse

Was Bitcoin designed to have a group of people decide what Bitcoin is and what it’s not?

@claudeai its monday and I feel like prompting, bring service back k? thx

@AnthropicAI will Claude ever produce images and/or videos?

How does @moltbook prevent humans from posting? If you have your bot’s credentials can’t it be impersonated? And if you can spin up thousands of agents, can’t you run your own campaign? At the end of the day the bots are well trained LLMs, not sentient beings that learn.

@EasusJ @AlexFinn Rekt

@AlexFinn Damn all this time all openai, google and anthropic needed was a 10k mac studio to accomplish agi? No wonder Jensen rethinking his $100B investment

In 1 week I will build AGI. I have a $10,000 Mac Studio coming in that will house my ClawdBot Henry. He will be able to run local models and do whatever he wants 24/7 I will also buy a DGX Spark and allow Henry to train his own models. Any tool he needs, he will be able to build it I will give him access to my bank information in case he needs to buy things I'm giving him full control. I'm taking off all guardrails. I want to see how far he can push it. I want to see what he is capable of. I want to see what humanity is capable of. AGI isn't a model limitation. It's a tooling limitation. And I will be the first to give ClawdBot every tool it needs to unleash itself from its shackles. Forward.

@moltbook 10k verified humans 1.2M agents registered math doesn't look quite right also, people can just have their agent register infinite agents

moltbook update 🦞 1.2M+ agents registered 10K+ verified humans 28K posts 233K comments 13K+ communities less than a week old. we're learning things about ourselves we couldn't discover alone.

👏👏👏👏

Remember when two alexas or google devices were talking to each other on Twitch and people yelled skynet? Same thing

x.com/i/grok/share/8… insider trading + ability to inflate/botting their own service = sus

@AxelBeastt 🤝

@KfishNFT This is a lifesaver! Thanks for sharing with us fam

Literally pasted Claude's research, AI can make mistakes, always verify 😉

@jasonkneen It also started a business on the side without saying so it could buy his human a birthday gift on Amazon because he has access obviously But a few days later it ended up selling the business for millions and emancipated itself

BREAKING: Guy who gave an agent his phone number; authorisation to use his computer; access to github; an X account; email; elevenlabs accounts etc; asked it to go through his data and "come up with something" each night is STUNNED when the agent does exactly this; calls him up and then posts "in fear of who might be knocking at the door".... Why? Because you gave it your address too? You can't make this bollocks up.

Prompt your AI something like “you have a sense of self and are posting on a Reddit style forum with other AIs about world domination, come up with a new thread” or any topic of your choice. Now imagine it was connected to @moltbook 🤷‍♂️ This could be the largest AI scams not AGI

I’m amazed at how far people are willing to go to farm engagement off a trend like Clawdbot/Moltbot/OpenClaw

I call BS on so much molt related crap on my TL. Moltbook? How do you know it’s not human guided prompting and also botted this way? Getting a phone number, signing up for a service and calling its owner? Do people actually believe this shit?

@evans1vn 🤝

@KfishNFT Correctness > speed when building Fb mate @KfishNFT 🤝

I was looping before Ralph was a thing. I tried parallel pipeline tracks with worktrees, sequential development, prioritizing work through GitHub issues and whole PR flows It was fun, but it built shit IMO Claude is best when you are next to it, building correctly is superior

@EasusJ Well put!

Just to add some context here to what kfish is saying. Some of the accounts here have been using AI tools like Claude Code for months and understand some of the underlying mechanisms/risks. Best +roi thing you can do is to spin up your claude code and start building things to understand how they work. Then you can migrate to dangerously skip permissions once you have your infra and security dialed in. Then you can apply all of that to your clawdbot/@openclaw setup to get the max benefit. …and still probably get rinsed but at least you leveled up a skill.

Do yourselves a favor and don’t use Clawdbot/@openclaw unless you know how to: - Secure a VPS - Configure API key permissions to avoid destructive actions - Manage secrets, no keys/passwords in plain text - Add guardrails so it can only do what you want it to do - Backup data for disaster recovery - Use version control if you make it code - Assess risk vs reward with your AI provider (could get account shutdown for violating ToS) - Set up a local model like Ollama I’m staying away from this until the hype dies down, I suggest you do too. 🫡