GFS doesn't compete against TSMC because they don't manufacture on small nodes critical to AI chips. But if physical and edge AI become as huge as so many people are saying (even Jensen), robots need many larger chips. I see GFS also just made a partnership with MRVL to manufacture photonics equipment. And the US gov't just made a 375M grant to GFS to develop an approach-agnostic quantum manufacturing capability. In addition to their core business, that seems like a nice, diversified portfolio for the future. I'm looking at them in case they re-rate. At the moment, they don't seem sexy; but that's why they may be a good purchase now for a long-term hold. GFS should also benefit if the US continues to favor US-based manufacturing for semiconductors.
Many of the companies that will supply semis to robotics seem to have their own fabs. But that might make GFS valuable for fabless chip designers. Does GFS have multibagger potential? I like that they only have a market cap of 25B right now but I don't know what else would come into play for a multibagger thesis. I imagine most redditors hate GFS because they're primarily pushing What's Hot Now (e.g. NBIS / CRWV etc) rather than looking further in the future; possibly professional analysts as well.