Anthropic's IPO Could Value It at 30 Times Revenue. Here's Why I'm Still Interested Anyway.

Anthropic is reportedly targeting a ~$2T IPO valuation, which would imply ~30x its $65B annualized revenue run-rate (vs. ~$9B at end-2025). The article argues the business may be priced expensively for now, but projects 2028 revenue of ~$190B–$200B—potentially placing the valuation at ~10x forward 2028 sales if growth continues. Sentiment is cautious-to-upbeat: the piece suggests waiting for an IPO pullback to buy rather than chasing the hype, citing only 9 of 36 similarly large recent IPOs outperforming the S&P 500 post-listing.

Analysis

A headline-grabbing private-market valuation is less useful as a standalone signal than as a fresh reference point for AI capital intensity. If the filing/marketing range is anywhere near the rumor mill, the near-term effect is likely to widen dispersion inside AI: infrastructure names with visible spend capture and recurring demand should keep their premium, while high-multiple software names with less measurable monetization become more vulnerable to multiple scrutiny.

The second-order risk is supply. A giant IPO creates a new liquid benchmark, but it also unlocks employee/early-investor distribution and invites a wave of late-stage AI markups to be re-traded in public. That usually helps the first 1-2 sessions and then shifts the market’s attention to growth durability, gross margin quality, and customer concentration. For public comps, the real read-through is not that the whole AI stack should re-rate, but that the market will become less tolerant of narratives that cannot show operating leverage.

Contrarian view: the consensus is likely over-indexing on absolute revenue growth and underweighting the fact that the IPO window itself is the catalyst, not the destination. If the S-1 reveals consumption volatility, concentration among a few enterprise accounts, or heavy compute dependence, the multiple could compress quickly despite strong top-line optics. The thesis is falsified if AI capex remains accelerating into the next earnings season and NVDA/NET keep converting that demand into guide raises rather than just sympathy rallies.

Key Decisions for Investors

• No direct chase in the IPO name until S-1 disclosure; wait for customer concentration, gross margin, and retention data before underwriting a valuation anchor.
• Relative-value trade for the next 1-3 months: long NVDA / short PLTR. NVDA owns the durable capex line item; PLTR is more exposed to multiple compression if the market starts pricing AI software against a fresh private-market benchmark. Falsify if PLTR re-accelerates growth and billings while NVDA guide momentum stalls.
• If the IPO tape triggers an AI beta selloff, buy NET on a 10-15% drawdown only if there is no company-specific miss. It is a cleaner enterprise AI beneficiary than narrative-heavy software, but the entry should be price-driven, not emotional.
• Set an alert around the IPO price range and first secondary sale. If the float is small and lockup overhang looks large, expect post-IPO weakness after the initial pop; that is the higher-probability entry point than the print itself.