We'll be spending a lot more time trying to understand the outputs of language models. A few thoughts, tips & tricks: Writing. Something I've had success with: Ask your LLM to explain something in ASD-STE100, it's a controlled language specification originally developed for aerospace maintenance documentation. LLMs well-versed in this language and it comes with heavy constraints on clean writing style that I often find a lot more readable. Sometimes I've tried to soften it a bit e.g. ask for "80% of the way to ASD-STE100" because the spec is quite stringent. But even better: Diagrams / images. Instead of writing, ask your LLM to create a diagram. These can be a lot easier to process, parse, and understand. But even better: Web pages. Ask for output "in HTML" to get a beautiful, interactive webpage. LLMs are getting really good at frontend and can create beautiful experiences, animations, etc. But even better: Explainer videos. The output format I am most bullish on is fully custom / bespoke explainer videos generated on any arbitrary topic. Experiment with things like "Create a 3b1b style video explainer on X. Use my ElevenLabs API key for audio narration". (you'd need an API key for the latter or you can ask your LLM to find you decent free alternatives that use your local compute). This is actually starting to work! In summary: - As LLMs get better, they will do more and more of the legwork autonomously, and a lot more of our work will rise up the abstractions into oversight and understanding. - Luckily, LLMs can help here too because as intelligence and code are increasingly abundant, you can ask for large, custom, discardable software artifacts (e.g. web apps, video explainers) that would have never made sense to create before. Push the boundaries here and you'll be surprised.

Try Grok @Bot

Grok Bot could be the first thing you hire instead of prompt It can own a job, work inside your tools for hours and come back with finished work In this article, I show you how https://x.com/i/article/2105262478535847936

Grok 4.7 writes great code.

I was a terminal person for 15+ years. I loved Vim, knew all the shortcuts, and a black terminal made me feel like I was a cool hacker in The Matrix. But terminals assume humans operate computers directly via files, commands, processes. Agents changed all that. You just say your intent. The agent operates the machine, using the same programming languages and Linux commands we spent years learning. I’m still glad I learned computer systems before the AI era though. Understanding what sits beneath the abstraction makes you a better systems thinker. That’s a moat in the AI era.

Gave Claude another 12 hours to one up this.. Woke up to.. The Clodyssey

Grok Bot is operating so insanely fast these days, it feels like SpaceX already launched AI data centers into orbit. Seriously… what is going on?! It’s gotten ridiculously good…

@davis7 6.1 coming soon

I haven’t touched Claude Code or Codex CLI in a while. The terminal era is over imo. It's the wrong interface for coding agents. Tabs are ephemeral, but context is persistent, and managing 30 tabs is pure cognitive overhead. I don’t really need an IDE like Cursor either. I rarely navigate the whole codebase anymore. The new primitive is the agent, not the file. (Codex desktop app is the best agentic UI for now. But we’re still early.)

I haven’t touched Claude Code or Codex CLI in a while. The terminal era is over imo. It's the wrong interface for coding agents. Tabs are ephemeral, but context is persistent, and managing 30 tabs is pure cognitive overhead. I don’t really need an IDE like Cursor either. I rarely navigate the whole codebase anymore. The new primitive is the agent, not the file. (Codex desktop app is the best agentic UI for now. But we’re still early.)

Jason Oppenheim, founder of Oppenheim Group ($5B+ in real estate sales), just sold his Bentley for a @Tesla Model Y and is so blown away by FSD that he’s buying Teslas for 10 employees: "This is the most important video I've ever posted. Tesla's self-driving is life-changing. It's so safe. I was on the phone with my brother last night, and I made him buy one. He literally bought one while we were talking. I'm buying 10 of my employees a Tesla with FSD. It's 8x safer than the average driver. There's nothing more important than the safety of you and your loved ones. Let's make our communities safer. I hope you'll at least consider going and getting a test drive." I've been saying for a couple months now that FSD has become a material driver of Tesla sales in North America. Some people said it's just been anecdotal evidence, but more and more videos and stories like this are popping up. It's not anecdotal anymore. Word is spreading fast. FSD is going mainstream. Jason's full video linked in thread below.

A humanoid is a capital good on a scale that changes the constraint. It lengthens production and raises what labor and land can turn out, which is the abundance Elon is pointing at. A person supplies about 2,000 hours a year. A robot on a 168 hour week supplies about 8,700. His floor of 1 billion within 10 years, priced in the low 10,000s, equals about 4 billion full time equivalents against a world labor force of 3.5 billion. The 10 billion at 15 years is an order of magnitude past that. Robots building robots still require saving, because energy, materials, and machines have to be bid into that line. Expected abundance holds for ordinary wants which is what matters for people as consumers. Housing, food, care, and routine medicine get repriced toward energy and materials until a household’s budget stops being what limits them. An owner running 100s or 1,000s of agents earns the return on that capital stock, which is also how universal high income shows up as purchasing power. Money prices remain the way to allocate whatever is still scarce, that is energy, locations, raw materials. At least in the short term. Energy is the binding input, and settling more than one planet adds land, minerals, and generation beyond a single world. TLDR: (1) More energy. (2) Multiplanetary doesn’t hurt.

Claude Ultracode is super intelligence.