What's Due For a "ChatGPT Moment?"

Hello folks! Hope you have great holiday plans lined up. Relatively new mid-level in venture / early growth investing and am starting to form my perspectives on investable themes that might emerge over the next 2 - 4 years. I'm still relatively young and the "COVID bubble" was my the first "economic cycle" of my career (I saw the run-up from a PE seat and then sat on the sidelines for a few years in grad school); having said that, it increasingly feels like venture returns are largely driven by large trends (market beta) in short windows of opportunity.

This begs the question - what are some themes that the collective wisdom of WSO believe will be interesting over the next cycle that haven't gotten heat yet? For example, it would have been nice to have been investing in AI in the late 2010's, or in vertical SaaS + fintech in the early 2010's. I have a few brief thoughts, but would be interested to hear where peoples' heads are at:
 

• Automation / digitization of the physical world: manufacturing, oil & gas, warehouse, and chemicals end markets (among others) have always seemed perpetually underpenetrated from a tech perspective. It feels like there are good reasons for this (high performance requirements, fragmented supply chain, tough customer base, etc.), but the opportunity to reduce downtime, increase efficiency, relieve labor bottlenecks, etc. seems so compelling and has so much broad market support that you would think (hope?) that this is an investable trend over the next 10 - 20 years.
• AI applications in biology and chemistry: I'm admittedly not a scientist, but it feels like the combination of expanding compute power (e.g., NVDA GPUs) and transformer architecture / LLMs should vastly expand computational potential in the hard sciences. I've always struggled with access to high quality training data (the people who would own the data, including pharma and CROs, don't have great incentives to share their information), but I suspect that there are business models that could unlock this data (though they might be highly specialized solutions rather than generalized solutions across an entire industry - TBD).
• Quantum: I'll admit that I really don't know what I'm talking about here. But I would be willing to bet that the first enterprise grade quantum computer will create a ten's of trillions enterprise value company someday.
• Edtech: Nobody has cracked the puzzle here, but teachers are in short supply, are underpaid, and are overworked; could the combination of LLMs, chatbots, and generative AI create new companies that teach students how to tackle higher-level thinking challenges such as word problems in math, improve their writing, and take the burden off of teachers in terms of personalized content generation, grading, and learning improvement plans? There's genuine potential for decacorns in this space, though I wonder if incumbents (e.g., content publishers, Google) might capture a lot of this value instead of new challengers.

Would be curious for thoughts on what people are excited about (and if these themes are nonsensical)!