IPO Fireworks Revisited

When we published Beyond the Numbers: IPOs in 2026, Prepare for Fireworks in June, our message was deliberately balanced. An unusually consequential initial public offering (IPO) calendar could give public investors access to some of the world’s most innovative private companies, but excitement around a business should not be confused with an attractive entry point. Our historical analysis reinforced this. We studied roughly 1,500 IPOs over the preceding 30 years and found that the average first-year return was 10.4%, while the median was -4.7%, and the average/median maximum first-year drawdown was 49/48%.

SpaceX (SPCX) has provided a timely example. In June, we specifically highlighted the potential for significant volatility following the SPCX IPO, emphasizing valuation, limited public float, lockups, and the difference between the IPO price and where ordinary investors can actually purchase shares. SpaceX priced its IPO at $135, and opened at $150, but the shares subsequently traded across a remarkably wide $104.83–225.64 range. On October 30, SPCX closed at $134.72, below its IPO price and its first public trade.

Now another highly anticipated company is providing investors with a window into a previously private corner of the AI ecosystem. Anthropic filed its S-1 on October 26, filed an amendment on November 2, and plans to trade on Nasdaq under the ticker ANT, although the initial filings did not specify the number of shares or anticipated offering price.

The filing is perhaps most useful as a primer on the economics of frontier AI. Anthropic generated just $41 million of revenue in 2023, followed by $386 million in 2024, $4.6 billion in 2025, and $33.2 billion in the first three quarters of 2026. As of September, more than 11,900 customers generated at least $100,000 of trailing 12-month revenue, including more than 1,900 above $1 million. Compute and infrastructure costs remain central to the business model, while Anthropic also identifies competition, regulation, customer concentration, model safety, and governance as meaningful risks. And while the breakneck pace of revenue growth is impressive, so too is the expected cost to support continued growth, as the company disclosed nearly $900 billion of compute commitments in their filings.

Another learning from the filing we find noteworthy is the company’s corporate structure. Anthropic is a public benefit corporation and has established a Long-Term Benefit Trust (LTBT) independent body that is tasked with holding Anthropic to its long-term public benefit mission. In their S-1, the company outlined how management behavior may differ from that of a purely profit-maximizing company. Notably with respect to how aggressively they will seek to secure compute as well as potentially delaying releases until some certainty can be achieved that the newly released model will not cause undue harm.

For investors, the framework we laid out in June remains unchanged: separate the business from the stock. Strategic importance, long-term economics, and the price public investors ultimately pay are three different questions. Anthropic’s filing gives markets substantially more information with which to answer the first two. The third will ultimately be determined by the public market itself.

LPL Financial does not offer access to or purchase of initial public offerings (IPOs).

This material is intended for informational and educational purposes only and does not constitute investment research, a research report, or a recommendation regarding any specific security or issuer.