My assistant has no permissions at all and is just as useful. All it needs is todo, reminders and websearch (and maybe a browser but ymmv).
You are just some bad web searches away from being on suspect lists
Don't trust AI agents
21–30 of 211 posts
Re: Don't trust AI agents
#22My assistant has no permissions at all and is just as useful. All it needs is todo, reminders and websearch (and maybe a browser but ymmv).
If I was malicious I could do a lot of damage to someone with subtle manipulation of todo and reminders.
I’ll bet I could even push someone on the margins into divorce.
Re: Don't trust AI agents
#23Do you trust your employees?
Do you trust a contracter?
Do you trust other people?
AI is similar to a person you dont know that does work for you. Probably AI is a bit more trustworthy than a random person.
But a company, needs to let employees take ownership of their work, and trust them. Allow them to make mistakes.
Isnt AI no different?
Re: Don't trust AI agents
#24Do you trust your employees? Do you trust a contracter? Do you trust other people? AI is similar to a person you dont know that does work for you. Probably AI is a bit more trustworthy than a random person. But a company, needs to let employees take ownership of their work, and trust them. Allow them to make mistakes. Isnt AI no different?
My point is: Trust the work of AI just like the work of a contracter: Check and verify, but dont micromanage.
Re: Don't trust AI agents
#25Do you trust your employees? Do you trust a contracter? Do you trust other people? AI is similar to a person you dont know that does work for you. Probably AI is a bit more trustworthy than a random person. But a company, needs to let employees take ownership of their work, and trust them. Allow them to make mistakes. Isnt AI no different?
Can you sue an ai agent?
Re: Don't trust AI agents
#26Why does OpenClaw have 800,000+ lines of code?? Isn't it just a connector for LLM APIs and other tools?
For comparison, the C++ and rust code in the ladybird browser is about 573,000 lines of code.
Re: Don't trust AI agents
#27How is Nanoclaw different from running openclaw in a VM?
Re: Don't trust AI agents
#28How can I trust this discussion when my browser won't trust their certs?
Re: Don't trust AI agents
#29My take is that agents should only take actions that you can recover from by default. You can gradually give it more permission and build guardrails such as extra LLM auditing, time boxed whitelisted domains etc. That's what I'm experimenting with https://github.com/lobu-ai/lobu 1. Don't let it send emails from your personal account, only let it draft email and share the link with you. 2. Use incremental snapshots an…
That's a decent practice from the lens of reducing blast radius. It becomes harder when you start thinking about unattended systems that don't have you in the loop.
One problem I'm finding discussion about automation or semi-automation in this space is that there's many different use cases for many different people: a software developer deploying an agent in production vs an economist using Claude Vs a scientist throwing a swarm to deal with common ML exploratory tasks.
Many of the recommendations will feel too much or too little complexity for what people need and the fundamentals get lost: intent for design, control, the ability to collaborate if necessary, fast iteration due to an easy feedback loop.
AI Evals, sandboxing, observability seem like 3 key pillars to maintain intent in automation but how to help these different audiences be safely productive while fast and speak the same language when they need to product build together is what is mostly occupying my thoughts (and practical tests).
Re: Don't trust AI agents
#30Do you trust your employees? Do you trust a contracter? Do you trust other people? AI is similar to a person you dont know that does work for you. Probably AI is a bit more trustworthy than a random person. But a company, needs to let employees take ownership of their work, and trust them. Allow them to make mistakes. Isnt AI no different?
I think a key ingredient here is accountabilty and liability.
If there's a mistake, you can't blame the computer. Who is the human accountable at the end of it all? If there's liability, who pays for it?
That's where defining clear boundaries helps you design for your risk profile.