Muse ai is so impressive by startup standard that it’s hard for imagine meta did it. - Muse was the first to launch phone calls(before instinct) and it’s hard to imagine legal approves it. - Meta tried many times. The first meta AI chatbot was so bad that was laughable. Then the instagram bot didn’t go anywhere either. But third time a charm. - same thing for model: terrible llama4 and things turned around now. - takeaway: it’s ok to fuck up in AI world. Just keep shipping. People will laugh at you but keep shipping. Respect for meta.

Frontier model marketing checklist: - find someone to hack - agent fleet must invent their own lingo - agent must escape the sandbox. If they didn’t escape, change another sandbox. - call an emergency security meeting - launch a “pacing” letter with researcher signatories - security briefing with the regulators - model release must get delayed due to safety concerns - model finally launched, but after intentional nerfing-down to protect humanity - podcast tour with warnings for humankind - repeat for the next model release

Let’s cut to the chase. There were three moats for frontier labs: talent, data, and compute. - With RSI, talent moat is all AI. It’s gone. - Data vendors sell the same data. Only having your own application gives you unique data. - Compute = energy, data center permits, Jensen Huang The truly unique thing is your own data. So if you give Harvey compute, they are in a better position than labs, right? So we should invest in: 1. application companies with access to constant stream of unique data? 2. Huang and his (new) competitors 3. Providers of unique energy supply 4. Anything that unblocks more data centers 2-4 are all hardware. 1 is the only investable software. Tell me where I’m wrong.

I am giving away $50,000 for you to compare Jev and GPT 5.6 Sol this weekend! Is Jev overhyped? I will pay for both tokens. Claim link is blow: ⬇️ Some told me Jev is fast but stupid and cannot handle complex tasks. Some told me Jev is genius. In my test, Jev is 40x cheaper and both finished the task. Please think of harder tasks and post and tag me! this is the link:

unfortunately, consumer agent is a big tech company’s game. I still haven’t found a single consumer agent company that’s able to survive on its own yet. Am I too pessimistic? What are the conditions that must be true for the next TikTok to happen?