hunether

There are two types of crypto AI projects currently where ppl can believe high valuations and they can have sustained HTF pumps. Agentic infra. These could be used by many other agents. E.g. $VIRTUAL, $TIBBIR, $FAI, $REPPO Model improvements. LLM’s are fairly limited. Memory is a probem and they can’t really learn or reason. The projects that try to improve LLM limitations could be used by many protocols/teams as brain. E.g. $REI, $VENICE, $ODAI The tickers are just examples. They might not go to insane valuations ofc but at least ppl can put bigger price tag on coins like these on HTF. On the other hand, it is much harder to justify insane valuations for agents without some insane and very specific narrative. Why would $CLAWD, $LOBSTAR, or $FELIX go way higher in the long run? I am not saying they can’t but there must be a very good reason (and a bit of agentic revenue will not cut it on HTF). Also the agentic meta/AI is changing quickly, so many prev agents will easily be replaced. Much harder to replace infra/model improvement projects where teams have been building for long months/years.

Let me get this straight: Anthropic refused to work with DoW unless they could promise their tech wasn't used for surveillance or killing. DoW said that they need full capabilities. Anthropic declined to give full access. OpenAI stood by Anthropic for ensuring AI safety. Trump then cancelled all Anthropic usage across the government, including a $200m contract. OpenAI then submits a bid to replace Anthropic.

Claude Code on desktop can now preview your running apps, review your code, and handle CI failures and PRs in the background. Here’s what's new:

Vibe coding is the new product management. Training and tuning models is the new coding.