1,421 karma · joined September 13, 2019
If anything, regulation is the more realistic solution. Requiring licensed CS/CE/STEM engineers (a degree plus a licensing exam, likely one for the US and a separate one for the EU) to sign off either manually written or generated by AI output on software in critical areas like healthcare, finance, insurance, energy, government and infrastructure, or above a certain revenue or valuation. That would keep people with verified skills in demand and improve security. Anyone could still build yet another SaaS, custom apps or games, and teach themselves along the way. They'd just need to hire licensed specialists once they grow or handle sensitive data.
It's far from a perfect solution, but an even stricter system already works in law, medicine and civil engineering. It just moves us from the current Wild West back toward the middle. Getting there would take real consensus, though, and tech workers have historically resisted organizing, since many see themselves as individual high earners who don't need a union. Add lobbying from large corporations, and it's a long shot.
Law and accountability is for poor people apparently, power is for rich.
Oh please no! I loved the old boring interiors with analog components
Yeah, that’s my thoughts as well. I feel it’s great for benchmarks and some tasks while in other it tries to spend as much tokens as possible, tries to overcomplicate task and needs seconds or third round of steering that costs. With the scale Anthropic operates I bet it’s huge amount of extra money just to make sure their model works.