mcintyre1994

I suspect that whatever Musk has agreed to spend on the midterms in exchange for those sweetheart deals the SpaceX IPO got will make that look absolutely comical, but we’ll see.

I find this weird too! When you say something that prompts them that you’re talking about the current time, they’ll go and fetch it and use that to understand time frames which makes sense. But if you’ve been talking about something tomorrow, and then tomorrow comes and you go back to that chat, they don’t understand that now it’s today/right now.

> Let's jump out of the middle of checkout and go to a page which looks like they're trying to phish your bank account to continue, and by the way, you can just skip it.

Any idea why they implemented it like that?

In the UK you either get a text from your bank with transaction details and a code to approve it. Or the better ones you just open their app and tap approve.

Fly.io’s sprites.dev is a similar idea but on their hardware and you get a public/private domain to run a server too, I’ve found it pretty good.

Same, the main network I see bots on is X and there’s a really obvious bot attack on them there since the fight with the DoD.

> We asked it [GPT-5.5] to assess whether each leader had made a falsifiable claim about the future as part of its main thesis. About 1,400 did. We then extracted those predictions, and asked the AI to mark out of ten both how contrarian the leader’s outlook was at the time and how accurate the prediction turned out to be. We ran those queries several times and took an average.

I understand why it wouldn’t be feasible for a human to do this, but I’m quite sceptical about an AI assessing how accurate predictions turned out to be/how contrarian they were at the time. It seems like that would depend a lot on what sources it chooses, be liable to hallucination or getting poisoned by bad sources, etc. They don’t mention whether they used independent queries for each prediction either, or whether it was doing multiple sequentially.

Given that LLMs can’t really distinguish prompt from instructions etc, I’m sceptical that they can reason particularly well about things like how contrarian a view was at a particular point in time.

I think I’ve mostly found superpowers helpful, especially for TDD. It’s cool from this blog (and their GitHub) that they’ve verified it in various ways too. One issue is that I’ll sometimes see it wanting to write a whole doc for a follow on feature with a similar structure and have to tell it to just use the last one again.

The most annoying thing is that it always pauses before implementing to ask if I want it to use Subagents (which it always recommends) or not.

I’ve definitely sometimes had situations where this would make sense, and also where Claude asks a question not using this tool and it’d make sense to just keep going.

But this does seem like the wrong default to me. I’ve found if I tell Claude I’m AFK and just keep going, it’ll do that reliably. And it’s good at debriefing me afterward on what decisions it made that it would have asked me about.

I have a really dumb prediction that at some point he’ll name one of the public companies X, presumably either SpaceX or a merged company. The dumbest people in the world would say he’s turned Twitter into a $2tn company and he’d find that hilarious.

Assuming you already pay for 'verification' on X, you just need to get more followers/impressions and then you'll start being paid based on your impressions. If you know/follow anyone with the 'verified' checkmark who has a lot of followers, they'll be getting paid for impressions.

It’d be restrictions on Americans and American companies, and probably also pressure on America’s allies.

OnePassword can do something like this where you put references to a path there instead of the key material, and then you wrap the invoke command with their CLI and it replaces them. So your local env file never has anything sensitive. A malicious agent could still exfiltrate if you give it access to debug tools on the running code though.

It's quite funny to think about the reaction this would be getting from people like David Sacks, if it was literally anybody except Trump.

Companies are spending even more than that if they’re using the $200 subscription worth of tokens on the enterprise plans too.

Supposedly the details of the ‘jailbreak’ are that you give it insecure code and say “fix this code”, and it does, and then you ask it for test scripts and that’s effectively an exploit against the unfixed code.

If true then I have no idea how anyone’s going to release a useful model that doesn’t have the same jailbreak. https://www.theregister.com/security/2026/06/15/feds-freaked...

I doubt there's enough evidence to make that claim yet, but I think 24/7 trading could do that.

I do agree that there's a huge amount of fraud and scams, and obviously that's only got worse since the President of the US started being part of that ecosystem.

But at the same time there is also finally real finance happening on-chain too. Backpack launched a SpaceX token at IPO that can be moved between on-chain and your brokerage. I think Coinbase announced their on-chain equity offering will have the same capability. Just yesterday Bailie Gifford launched a tokenised fund where the actual register of record is on-chain. I still think crypto has significant potential as financial rails, and that does seem to be being explored by real financial players now too.

I guess the idea is that you probably can, or will be able to, find a host that you trust at least as much as you trust Anthropic.

I find the same. Someone posted this benchmark here: https://petergpt.github.io/bullshit-benchmark/viewer/index.v...

It measures whether models push back on bullshit prompts or just go along with it, and Claude models are all the top performers.

A few years ago someone kept signing up for loads of bank accounts/credit cards in my name, with my address. I’m not sure what the point of it was. But while everyone else happily sent cards and stacks of welcome paperwork to me, Amex were the only one that contacted me and told me they’d detected something weird in the signup. They gave me some helpful advice to resolve that situation too.

I feel like it has to eventually become X. Just because he’d find it funny for the stupidest people in the world to post about how he’s turned Twitter into a $2t or whatever company.

It's all stock

TBH this is the main thing that made me start trusting Claude enough to actually find it useful, and I'm surprised other models haven't caught up. I assumed they had and I just wasn't aware because I'm not using them in the same way.

I think this is just Apple planning for their on-device models getting better, which makes sense given they have access to Gemini now. If developers use this for all their code calling an external LLM, then as Apple's model becomes more capable and covers more use cases it'll be easy to switch to it at individual call sites. That'll give apps better UX and save developers money on a bill that Apple doesn't get a cut of.

I haven’t used it for anything in production so I haven’t seen these issues, could you give a bit more detail? I’m mostly wondering why you’d have any more websocket messages that you respond to than you would APIs if you were using any other approach. Does LiveView itself respond to certain messages bypassing your app?

Ah okay thankyou! Is the MCP server manually configured, or is there documentation on the suggested way to tell an agent to use sem? My guess was that setup was how to do that.

This looks really neat, but I see its diff output as complimentary to git and wouldn’t want to replace git diff. Is there a way to just install the CLI and MCP server but not override git diff?

I’m guessing they have no functional human support for the people who had their accounts stolen. I get the impression Meta didn’t know this was happening until they were contacted by the media.

I get what you mean but SpaceX owns xAI, which is objectively a company that trains models and has massive distribution by owning X.

I don’t think their models are competitive with Google, and Google obviously has the best distribution imaginable, but they definitely are a competitor.

Arguably the old fashioned ones where the CEO doesn’t have special shares that give them a voting supermajority?