4 ms·
only in software there's a clear, tried and tested multiple times path for 1000x ROI for an investor. thermodynamics themselves limit ROI for anything physical.
by baq 1y ago
only in software there's a clear, tried and tested multiple times path for 1000x ROI for an investor. thermodynamics themselves limit ROI for anything physical.
- kridsdale3 1y agoThermodynamics is also the limiter in AI these days.
- hattmall 1y agoNot necessarily the real limiter, right now, is finding a combination useful and profitable application. Simply charging money for generalized AI access is never going to be the ideal profit center.
- 0xDEAFBEAD 1y ago>Simply charging money for generalized AI access is never going to be the ideal profit center. Exactly. Margins are dropping rapidly: https://ethanding.substack.com/p/openai-burns-the-boats https://ethanding.substack.com/p/openai-burns-the-boats
- baq 1y ago...and to quote a classic, "your margin is my opportunity".
- baq 1y agoIt limits token generation, but not token utility.
- athrowaway3z 1y agoAI is two things. - The chatbot people have a personal attachment to - The processing tool. In the second, you only care about the result. Something like Claude Code can call any other provider if that's cheaper and visa-verse. Once I have the result, my dependency / lock-in is no more than a brand of toilet paper. The providers will have to do the 'capitalism thing' and compete. It's almost like WeWork's, valuated at IT levels by being in the style, only for investors to eventually figure out the marginal production costs are not reducible to near 0, and you can't just bully out competition / network-effect to get a monopoly. And this applies to any company that wraps and re-sells AI. Something the tech-VC world is so unfamiliar with, it's scrambling to present the truth of what is 'econ-101' for the rest of the world.