> ## Content Index
> Fetch the complete content index at: https://wire.fourthweb.ai/llms.txt
> Use this file to discover other available public pages before exploring further.

# Howard Marks Says Nobody Can Value AI Companies Yet
- URL: https://wire.fourthweb.ai/howard-marks-says-nobody-can-value-ai-companies-yet/
- Published: 2026-11-02T15:43:47.000Z
- Updated: 2026-11-02T17:32:11.000Z
- Description: The smartest money in the room just admitted it can't price the future, and that should tell you something about everyone else pretending they can.
- Author: Travis Wright
- Tags: AI Agent Economy, AI Agents, Anthropic, IPO Watch, Funding Rounds, China AI, Big Tech

**The smartest money in the room just admitted it can't price the future, and that should tell you something about everyone else pretending they can.**

### The Summary

- [Howard Marks says traditional value investing breaks down with AI companies](https://www.businessinsider.com/howard-marks-ai-anthropic-stocks-nasdaq-investing-valuation-buffett-burry-2026-11?ref=wire.fourthweb.ai) because projecting earnings five years out is impossible when the product roadmap includes "constantly reinventing the business"
- [Recent AI market stumbles have "somewhat burst the bubble of infallibility"](https://www.bloomberg.com/news/articles/2026-11-02/oaktree-s-howard-marks-says-ai-stumbles-have-boosted-skepticism?ref=wire.fourthweb.ai) around the sector, per the Oaktree Capital co-chairman
- He used [Anthropic](https://wire.fourthweb.ai/tag/anthropic/)'s upcoming [IPO](https://wire.fourthweb.ai/tag/ipo-watch/) as the test case: can anyone honestly predict what Claude's parent company will be doing or earning in 2031?

### The Signal

[Howard Marks doesn't do hype](https://www.businessinsider.com/howard-marks-ai-anthropic-stocks-nasdaq-investing-valuation-buffett-burry-2026-11?ref=wire.fourthweb.ai). The billionaire value investor built Oaktree Capital on a simple premise: figure out what a company will earn five years from now, discount that back to present value, and buy when the market is selling below that number. It's the Warren Buffett playbook. It works when you're valuing railroads or insurance companies or even software businesses with predictable subscription revenue.

It doesn't work when the company's IPO prospectus explicitly says they plan to "dwarf the economic value of our existing ones" with products that don't exist yet. That's the [Anthropic situation](https://www.businessinsider.com/howard-marks-ai-anthropic-stocks-nasdaq-investing-valuation-buffett-burry-2026-11?ref=wire.fourthweb.ai). The company behind Claude is prepping what could be one of the biggest IPOs in history, and Marks is saying out loud what most investors won't: "Can we really fool ourselves into thinking we know what it will do five years from now?"

> "I don't think we can do that with much of AI."

The Anthropic prospectus gives you the timeline of why this is hard:

- 2022: Chatbots became viable
- 2025: Coding agents shipped
- 2026: Knowledge work agents and scientific discovery tools are just starting to work

What's the product in 2031? What's the margin structure? What's the competitive moat when the technology layer is moving this fast? Traditional valuation models need answers. Marks is saying those answers don't exist.

The other thing [Marks mentioned](https://www.bloomberg.com/news/articles/2026-11-02/oaktree-s-howard-marks-says-ai-stumbles-have-boosted-skepticism?ref=wire.fourthweb.ai): recent market stumbles have cracked the sector's "bubble of infallibility." He didn't specify which stumbles, but the implication is clear. The AI trade got ahead of itself. Reality checked in. Skepticism is back in the room.

Here's the paradox Marks didn't resolve: "And yet, that doesn't mean we don't want to be in it." He's acknowledging that the inability to value something doesn't mean it's not valuable. It just means the traditional tools don't apply. You're either comfortable operating in that uncertainty or you sit it out.

### The Implication

If Howard Marks can't value AI companies with his decades of experience and billionaire-grade patience, what does that mean for retail investors piling into AI stocks or the venture funds writing $100M checks at $30B valuations? It means most people are guessing. Some guesses will be spectacularly right. Most will be wrong. The difference won't be analysis, it will be timing and luck.

For builders and operators, this is actually good news. If the market can't price you accurately, you have more room to build without quarterly earnings pressure warping your roadmap. Anthropic's prospectus claiming they'll constantly reinvent the business isn't a bug, it's the strategy. The valuation uncertainty is a feature if you're long-term greedy.

### Sources

[Business Insider Tech](https://www.businessinsider.com/howard-marks-ai-anthropic-stocks-nasdaq-investing-valuation-buffett-burry-2026-11?ref=wire.fourthweb.ai) | [Bloomberg Tech](https://www.bloomberg.com/news/videos/2026-11-02/howard-marks-says-ai-stumbles-have-boosted-skepticism-video?ref=wire.fourthweb.ai)