rocketman

new jungle dropped and i diggg open.spotify.com/track/5SHtU2P8…

Pause AI rhetoric is predicated on the notion that the AI companies are recklessly racing toward dangerous tech and that a government controlled pause button is therefore necessary, but this seems really hard to reconcile with the fact that government is attempting to destroy an AI company because *the government* is racing toward plausibly dangerous AI uses (Sec. Hegseth has stated in official directives that he wants to deploy AI into critical systems regardless of whether it is aligned, for example) and *the company* is pushing back. The roles are totally reversed from the logic that Pause AI and frankly other AI safety advocates confidently assumed for years. It is *industry* that is in favor of alignment and at least somewhat measured deployment risks, and government whose actions seem much closer to reckless. I predicted this for years. I said, in particular, that pauses and bans and licensing regimes gave government a dangerously high degree of control over AI, and that the incentives of government are much more dangerous than those of private industry with competitive market incentives. I believe the events of the last month are good evidence in favor of my view. At this point if you are an AI safety advocate whose policy proposals do not wrestle seriously with the brutal political economic reality of the state and AI, I don’t take you seriously. It gives me no pleasure to have been right about this, by the way. The state has an incredibly strong structural incentive to centralize power using AI, and we are, all of us, not so empowered to stop it. I am quite concerned about this.

@wevm_dev elite

Introducing webauthx A minimal WebAuthn toolkit for passkey registration and authentication, with conventional defaults for broad compatibility. 3 functions to represent the auth ceremony across server and client. npm i webauthx npx skills add wevm/webauthx

looks like sherlock.xyz is moving in this direction

wild you can get near unlimited Solidity protocol audits for $200/mo GPT Pro subscription that are on par with a $20k+ contest on code4rena. still feels like there's room to improve, any auditing teams moving AI-first?

at what level of intelligence does reinforcement learning become morally abhorrent

@emollick ignores the fact that limitless software can likely automate most jobs

Yes, this is the competitive advantage argument economists have been discussing for awhile, but it is coming true. There will not be enough compute for many years to automate many human jobs, even assuming AI can do that work and companies are willing to replace the people.

A big determinant of AI's job impact is driven by the lack of compute, especially for agentic work, which takes a lot of it. That makes AI expensive. So companies will only want to burn compute on high-value tasks (eg coding), because, in other jobs, humans remain much cheaper.

another day, another codex cli 5h limit hit

@polynoamial "But we need laws, not promises, to constrain this kind of power." imo laws are promises. really need hard tech guarantees NSA et. al have broken countless laws before. better approach to make it technically impossible vs hoping the police will police themselves

@paulg I like this one a lot! Curious if you've read subpixel.space/entries/life-a… from @tobyshorin - very similar and you might get some more ideas

The Brand Age: paulgraham.com/brandage.html

@antoniogm for ERC-4337 smart accounts (CDP sendUserOperation), does Base attribution require the suffix at the very end of final userOp.callData, or is appending to inner calls[i].data also indexed? We can't currently control raw userOp.callData directly. cc @programmer

@romuliiwww if you're trying to make websites maybe claude code or v0.app anything else maybe the codex app from @openai

@rocketman_w Hey bro. trying to get started w/ AI vibecoding. so many models Im at an impasse of which to choose and how to start. Should I start with 5.3 or Claude? What environment do u suggest? gravity, studio, windsurf, replit etc? Which is more economical for a starter? Or qwen3.5 oss?

Codex 5.3 is fucking unbelievable I'm paying 4x for AI vs last month, rest of the world is gonna 10x their spending when it crosses some usefulness threshold in their daily lives

@Paul_Glavin @thefireflyapp great app

(pay no mind, getting set up on @thefireflyapp)

gm 🌱

claude feels sloppy in comparison

feel like tony stark talking to JARVIS with all of these codex agents man

the cobuild agent is starting to get it

building a CLI for agents to use our protocol and issue wallets via CDP server wallets big bet is they'll pay using x402, but if we get charged $0.005 per signature on a server wallet, that kinda defeats the purpose of micro buying API calls and doesn't seem feasible for us to support CDP for agent wallets signing x402 payloads under that cost regime tbh. if we use the EOA offering instead of smart wallet, can we somehow get this per signature cost waived? or do you all have plans to revise this cost structure soon?

100

@tszzl @heyellieday @UnderSecretaryF @DeptofWar @kyliebytes so safety harness > actual redline in the contract? because the legalese in the OAI contract doesn't explicitly forbid autonomous weapons at all tbh. curious wdyt

@heyellieday @UnderSecretaryF @DeptofWar @kyliebytes my (limited) understanding is that openai gets to build a safety and monitoring harness that prevents unlawful uses especially in the important categories

For the avoidance of doubt, the OpenAI - @DeptofWar contract flows from the touchstone of “all lawful use” that DoW has rightfully insisted upon & xAI agreed to. But as Sam explained, it references certain existing legal authorities and includes certain mutually agreed upon safety mechanisms. This, again, is a compromise that Anthropic was offered, and rejected. Even if the substantive issues are the same there is a huge difference between (1) memorializing specific safety concerns by reference to particular legal and policy authorities, which are products of our constitutional and political system, and (2) insisting upon a set of prudential constraints subject to the interpretation of a private company and CEO. As we have been saying, the question is fundamental—who decides these weighty questions? Approach (1), accepted by OAI, references laws and thus appropriately vests those questions in our democratic system. Approach (2) unacceptably vests those questions in a single unaccountable CEO who would usurp sovereign control of our most sensitive systems. It is a great day for both America’s national security and AI leadership that two of our leading labs, OAI and xAI have reached the patriotic and correct answer here 🇺🇸