This week on The Learning Corner by Precursor, Mia and Charles dive into three major topics shaping the future of AI, startups, and software economics: Taco Bell’s Voice AI Troubles: A WSJ piece reveals just how glitchy the chain’s drive-thru AI rollout has been—and why they’re rethinking it entirely. Charles shares why these failures might actually be a good sign of progress and what’s at stake when AI misfires in higher-risk industries like healthcare and finance. The Myth of “Non-Consensus” Investing: A provocative thread from Martin Casado questions whether early-stage investors are placing too much faith in being contrarian. Is alpha really found in non-consensus bets—or does follow-on capital always chase what’s hot? The Software Margin Cliff: A brilliant Substack from Sam Schillace suggests that AI is driving software economics off a cliff. As inference costs grow and traditional SaaS margins shrink, the industry may look more like manufacturing than the creative playground it used to be.

(0:18) Introduction to the Learning Corner podcast (0:48) Taco Bell's bumpy rollout of voice AI and its broader implications (3:19) AI in healthcare, fintech, and the future of engineering (6:09) The nuances of contrarian and nonconsensus investing (10:09) Consensus in later-stage investing (12:46) Sam Schillis on AI's impact on software economics and changing business models (18:36) Closing remarks and thank yous

Episode #46: AI Troubles for Taco Bell, Is Non-Consensus Dangerous, Are Software Economics Going Off a Cliff with AI