Actual AGI might be really bad but thankfully we are not even close to building it and I doubt I’ll see it in my lifetime (imo ofc)

▲

anon84873628 18 hours ago | parent | next [-]

And even then, reaching "I, Robot" levels is unrealistic. Machines are fragile. It would take a very long time to bootstrap the entire automated robot construction and maintenance supply chain, which would also require energy production and resource collection too.

Until then it is vulnerable to disruption. So of course you will need the robot army to forcibly take and defend resources around the world. How will that war turn out?

Robot labor will be more valuable than humans for a very long time. Heck, we still have many towns in the US that have humans running around throwing trash bags into the back of the truck, because they haven't switched to the standardized containers and hydrologic arms. Let's not even get started on the rest of the "developing" world.

	▲	OgsyedIE 18 hours ago \| parent [-]
		Any non-maritime part of the supply chain can be extremely well-defended from disruption as long as you marry agentic defense planning based on existing crime prediction machine learning SaaS platforms to loyal human puppets with boots and guns (police, ICE, military).

▲

hackinthebochs 17 hours ago | parent | prev [-]

If you're betting the farm on there being a wall on the current trajectory of LLMs, I suggest you hedge that bet.

	▲	slopinthebag 15 hours ago \| parent \| next [-]
		If I wasn’t betting the farm I’d go into debt arming myself somewhere in northern Siberia with enough rations to live a lifetime. Since neither of us are doing this, we’re apparently both betting our lives on the current trajectory of LLMs hitting a wall soon and fast.
	▲	VCFundedGenYer 17 hours ago \| parent \| prev [-]
		LLMs are very limited and have finite usefulness. The idea of AGI is a grift.