Let's get this straight. The poster wants the software running on ARM and a new 20-core Intel chip. The software has to output flash and native GUI (on what platform(s)?), parse RTF and SVG and transfer data over FTP. It has to process gigabytes of data quickly, in limited memory.
I probably wouldn't write such a project using just a single language or set of tools. It sounds like there's a data-processing component, a mobile app, a desktop app and a web interface. I wouldn't pick a language or tool set for any of those components before getting detailed requirements.
I think the author is trying to clarify how much we really trust certain languages. But the $100M just muddies the issue. $100M, appropriately spent, could get almost any language across the finish line -- as long as you're willing to fork the language itself or add custom lower-level extensions. All the large companies that use high-level languages do this; Google, Yahoo, ITA, everybody.
Any serious attempt for the prize would have to include DBMS and operating system. This was, in fact, exactly what happened during Viet Nam as TRW responded to the U.S. Army's open ended requirements with GIRLS in 1965.
That effort survives today as IBM's U2 product line and still one of the best prototyping environments ever developed. (I've seen 20 year old "prototypes" written in Pick.)
Problem was, there was no $100,000,000 prize. That kind of money later went to Microsoft for products with (let's just say) a little less inspiration.
Leaving people like you and me to ponder the question on boards like this.
I probably wouldn't write such a project using just a single language or set of tools. It sounds like there's a data-processing component, a mobile app, a desktop app and a web interface. I wouldn't pick a language or tool set for any of those components before getting detailed requirements.