Two AI hardware stories hit the tape at once today — and only one is the bet everyone thinks they’re making. If you’re just joining, Greater China’s AI listing rebound started as an issuance-and-demand story. Z.ai and MiniMax were among the names that surged after their IPOs, and that helped rebuild confidence in Hong Kong’s tech pipeline. But that strength came with a supply overhang: cornerstone and insider shares from those deals were set to become freely tradable once the lock-ups ended. This is AI IPO Watch. Today: Syntiant files an edge-chip IPO that’s nothing like the data-center trade, and Hong Kong’s lock-up clock finally hits zero. The dates I called aren’t hypothetical anymore. Syntiant first — because not every AI chip is the same trade. Let’s get into it. If you want to keep up with Greater China AI listing rebound, tap follow so the next episode lands in your feed. Here's TradingView:
Syntiant, a provider of ultra‑low‑power edge AI chips, sensors and software, filed for an initial public offering. The company plans to list its Class A common stock on the Nasdaq Global Market under the ticker “SYTN.” Proceeds will support general corporate purposes, R&D, and potential debt repayment or acquisitions.
Finally, something on this beat that isn’t a data-center bet. Syntiant filed to list on the Nasdaq Global Market as SYTN — ultra-low-power edge chips, the stuff running inside an earbud instead of phoning home to a cloud server. And it’s a Class A common stock filing, which tells me dual-class is coming. But here’s the thing: hardware margins are legible. You can’t fog up a chip business with ARR projections. There are fewer places to hide in this S-1. Which is exactly why I want to say the contrast plainly. Syntiant isn’t the same AI IPO trade as the data-center names. It ships sensors — over 25 billion SiSonic units cumulatively, per the filing. The bet here is volume, not scarcity. And Intel’s on the cap table as a backer. So a semiconductor incumbent has a direct stake in whether SYTN prints well. When the syndicate lands, read it — that’s where you’ll see how much ambition is priced into this thing. The Intel stake is the tell for me. Strategic backer, edge silicon, VC clock ticking — the timing reads more like exit pressure than market euphoria. South China Morning Post, with Zoe SL Chan:
Hong Kong’s stock market could face sell-off pressure amid a torrent of new share supply in coming days as the six-month lock-up period ends for hot artificial intelligence and semiconductor picks including Zhipu AI and MiniMax.
So SCMP calls today a gut check. I called this date on Monday, and it’s not sentiment theater — it’s the cleanest price discovery Hong Kong gets all year. The cornerstones who bought in at the IPO weren’t exactly price-sensitive. They took allocation for the relationship. Now they can actually sell, and we find out what the marginal buyer will really pay. US$11.5 billion of supply legally unlocked in one window, per SCMP. That’s the number that matters, Eric — not the reported valuations we opened the week arguing about. This is where the Greater China listing rebound runs into supply risk. Zhipu and MiniMax got to skip the sell pressure for six months. Today, the clock runs out and the tape answers. Right — the slide going in was front-running. Now we find out whether that was the whole move or just the opening act. Some of these firms are eyeing placements on top of it. So you’ve got the unlock and the issuer adding supply into the same window. Gut check understates it; call it a stress test. Here's The Motley Fool:
Since then, the stock has pulled back and is down to around $162. The stock remains above its $135 IPO price and is still far from cheap. But the sell-off has made the risk-reward question more interesting.
So the Motley Fool wants to call $162 a contrarian buy — down from that $225 intraday high in June. The thing I keep coming back to: today is the Nasdaq-100 inclusion date. The forced passive buying is happening right now. And the stock’s sliding into it. Retail’s going to read a soft open as the index trade failing. I read it differently — the price found its level before the robots even showed up. The contrarian case leans entirely on Starlink — 60% of $18.7 billion in revenue, $4.4 billion in operating income. Those are the numbers actually in the document. And right underneath it: a $4.94 billion net loss for 2025. So the profitable business inside the unprofitable company is funding the ambitions. The Fool frames the slide as opportunity. I’d frame it as the market repricing $135 up to $225, then admitting it overshot. Okay, step back for a second — when people talk about a lock-up expiration after an IPO, what actually gets unlocked, and who are these insiders we keep hearing about? Great question, because “insider” gets used pretty loosely. Investopedia defines a lock-up agreement as a contractual restriction that keeps company insiders — employees, early investors, and founders — from selling their shares for a set period after an IPO, usually 90 to 180 days. The point, per Investopedia, is to protect new public-market investors from a wave of insider selling that could crater the stock right out of the gate. These agreements aren’t federally mandated — the SEC doesn’t require them — but they’re basically universal in IPOs because underwriters usually demand them before they’ll take a company public. KB Financial Advisors puts it really simply: during the lock-up window, insiders can watch the stock move, but they can’t act on it. Once that window closes, all that pent-up supply can legally hit the market at once. And the key dynamic is expectations. As the expiration date gets close, investors start doing the math on how many shares could come loose — and that anticipated selling pressure can move the stock before a single insider share changes hands. So the stock can drop just from the fear of selling — even if most insiders decide to hold? Exactly — the possibility alone can move the stock, even before anyone sells. Investopedia flags this directly: stock prices may drop around lock-up expiration, and informed investors sometimes treat that dip as a potential buying opportunity if the underlying business is sound. So for listeners watching any of the big AI-era IPOs, circle the lock-up expiration date. The volatility window often opens before the date itself. Have a tip on an AI company heading toward the public markets, or a correction we should know about? Send feedback and story ideas to aiipowatch at lantern podcasts dot com. We read every note.
Now we’re watching the next few days of Hong Kong trading in Zhipu AI and MiniMax, as their six-month lock-up expirations feed new share supply into the market.
Links to every story we covered are in the show notes, so if one deserves a closer read, start there. That’s AI IPO Watch for today. This is a Lantern Podcast.