- The innate value of the models come from not training chips nor hosting, but the actual quality of thinking traces, inference costs and usage. Anthropic shines here, but not for long.

- The variety of skills and evals on new tiny open models proves Claude is expensive and not extremely better than these models today, and math benchmarks alone do not cut it.

- Enterprises are realizing that they can just buy some regular Macs and do their own inference far cheaper than invest in Claude Subscriptions, and Claude given tools are losing their edge to harnesses such as Pi and Opencode.

- Nobody needs Hacking Tools unless they are red-teaming their own code. The niche of people wanting this could write their own tools and workflows to test the effectiveness of many LLMs on these workflows and see how far they can push it.

- The amount of collusions, hypes and fear mongering do not sit well with many people (including myself) - stick to the science and make usage safe, like everything else we do with software in production.

- The IPOs often get priced down with market discovery and anyone who's buying it ONLY is interested in the long run, not inflated valuations not hype not any of these lies used to sell this stock. I am bearish about the IPO at this rate.

- For me, reasons for SpaceX money investments are NOT due to AI *That part is a true waste of money* but good enough due to manufacturing, supply chains and actually building useful Satellite infrastructure; I still am bearish about them launching NVL72 nodes in space, not because of the space part - I just don't trust those chips enough for it to sustain that long.

>>SpaceX, meanwhile, trades at 16 times that year’s revenue. Take that as the benchmark — they are both companies with wild aspirations and charismatic leaders — and Anthropic could in future be worth $5tn.

SpaceX has a humongous moat in significant part of their business: launch and Starlink. They are so far ahead other players barely register. There are a lot of options connected to that moat. They also are good at building things fast and already have good infrastructure and are in position to build more.

This is not the case at all with Anthropic. Anthropic is very slightly ahead right now and some of their competitors have structural edges that are hard to overcome. For example Google has data and infra, SpaceX has infra an some unique data, Meta has a lot of data, infra and a cash machine to finance it all and operate at loss as long as they please.

That I think is the biggest problem for Anthropic. There is very little moat and competition is already close and very motivated to catch-up.

If I had to bet they will find buyers at 2T. Very liquid prediction markets are at 2.2T right now. I think they face uphill battle to justify that valuation even if AI sector continue to grow fast though.