Hanson here appears to be backing up his claim with three hyperlinks. One thing that troubles me about rationalist-affiliated writing is that it tends to rest on what looks, at first glance, like a decent evidence base, but this evidence often breaks down under a little scrutiny. For example:
• Hanson's first link is to a recent National Geographic article that summarizes an archaeology finding from Bronze Age China. The summary is, IMO, misleading in one key respect. It calls the find "astonishing" because "parts" of the site "date back 4,300 years, nearly 2,000 years before the oldest section of the Great Wall—and 500 years before Chinese civilization took root on the Central Plains." The Great Wall reference is a non sequitur: it's a bit like saying that a find from the Old Kingdom of Egypt is astonishing because it dates back to two thousand years before the building of a Roman aqueduct. There's no real connection, and it's not at all news that agricultural civilizations existed in China circa 2300 BCE. Whether we want to call these societies "Chinese civilization" or not is debatable. But this finding, while really interesting and cool, is very different from what Hanson seems to be reading it as -- which is to say, as evidence for the belief that future discoveries might yield signs of "advanced civilizations" which far predate the Bronze Age as it's currently imagined. After all, in 2300 BCE, written records and urbanized societies already existed in a fairly large swathe of Eurasia.
• His second link is simply to the Wikipedia page for Göbekli Tepe. Which is a totally fascinating site, I agree. It certainly does have unique features, such as the megalithic carvings. But again, it is not evidence of an entirely unknown "advanced" civilization; instead, the people who built Göbekli Tepe clearly fit into the existing archaeological record relating to the Neolithic Near East. We've known for some time that early agricultural societies were emerging in the fertile crescent around the same time as this site. What's fascinating about this site is that it appears to predate agriculture in the region. However, it is thought to fit into a transitional stage ("Recent DNA analysis of modern domesticated wheat compared with wild wheat has shown that its DNA is closest in sequence to wild wheat found on Karaca Dağ 30 km [20 mi] away from the site, suggesting that this is where modern wheat was first domesticated"). So again, IMO this has nothing to do with his claim about "lost" civilizations which don't fit into our existing chronology.
• The final link is to a Nature write-up from two years ago relating to a contested claim about when humans arrived in the Americas. Again, this has nothing to do with evidence for an "advanced civilization" that has previously not been discovered.
All three of these links reference really interesting and important work which I very much hope continues to be funded and elaborated on. I'm convinced that there are incredible mysteries in store in the future, especially in the field of underwater archaeology. Yet to my reading, none of these reports relate to phenomena that scholars "confidently predicted would never be found." Any archaeologist or historian with a sense of humility will gladly admit that there are a huge number of unknowns in the historical record. None of this can be leveraged as a basis for the existence of "lost advanced civilizations" (by which he appears to mean agricultural or otherwise technologically-advanced civilization that far predate the Agricultural Revolution as its currently understood, and which were forgotten after some kind of hypothetical period of decline).
For that, you need what Hanson waves away: actual archaeological or historical evidence.