Thats what Elon meant when he said he would find out who was pulling the strings. It starts with following the political trail of corruption

People not realizing they're getting jabaited loll Me and you know whats up with our level 30 chars.

Who gives a fuck honestly TMZ???

In Forever you'll be able to do a quest (unlocks at level 14), that gives you a "cozy sleeping bag" item. Gives you increased experience % and you gain rested XP aswell. not sure of the amount of XP tho

yes, but to level faster don't forget to do the sleeping bag quest when Forever goes live

@grok whats the tldr of what he wrote?

Thought that was a whale

so basically if you hate someone, buy gold and mass mail each of them some gold and boom they get banned? flawed system.

Claude Sonnet 5.5 is now available:

In physics, an “impedance mismatch” occurs when two systems each work well but are poorly matched. In this Science Blog guest post, Harvard physicist Matthew Schwartz argues that something similar is happening with AI and science. LLMs are capable at many things, but working with them as you would with a human collaborator isn’t currently the best way to elicit their scientific strengths. To address this mismatch, Schwartz created a toolkit for exact calculations in quantitative science. Because similar calculations often emerge in very different areas of science, Claude found connections to ecology, population genetics, and a dozen other fields, and Schwartz worked with domain experts to steer it towards interesting questions. Read more about these projects here: anthropic.com/research/claud…

An agent that combines reliable computer use with strong intent understanding would be a meaningful step beyond today’s tools.

The comparison comes down to both model quality and product fit; different workloads can make the gap feel very different.

A $100 million training commitment is substantial; the 10,000-engineer target makes the deployment focus especially clear.

The unit economics are a useful reality check for computer-use products, especially as orchestration costs come down.

more intuitive*

A nice little QoL upgrade would be an actual intelligent live goal check while we work. It should infer what you're actually building and the longer horizon, then call out when the current work is fighting that goal or jumping the gun on steps that actually matter. This would help write less code down the wrong path and the session stays productive without wasting the user's tokens or time developing trash results due to poor thought out execution.

What’s one thing that’s missing in codex that you wish we had?

Get urself some Raid spray

Not while G2G is still a thing