siliconc0w

I think it'd be interesting for a firm to try to run without traditional software. Like the computer boots, you get a floating orb and you tell it stuff, it remembers, it stores it in an optimal format, it is capable of receiving, triaging, and sending synchronous or asynchronous communications externally or around the organization. It knows how to build just-in-time tools and can recall any information, displaying it in the most helpful format.

It's not clear what 'work' those orbs would be doing because a lot of 'work' exists because of current frictions due to bad tools or some arbitrage around a lack of trust or expertise. For example with contract redlines, if I just tell my AI what is important to me, and you tell yours, they can collaborate on a shared agreement that is essentially guaranteed to be fair (modulo trusting the AI). No emailing word documents back and forth, no ambiguity of obligations, no hiding misleading terms, etc.

I kinda like that it existed. It was ridiculous, like a production concept car. I just saw one the other day with a pink wrap.

It's pretty crazy this is tolerated by society, I know several older people scammed using Metas products.

And that isn't even counting the Instagram ads my wife falls for.

Pretty boring with ChatGPT Pro, ghostty, CLI harnesses, and vscode if I feel like looking at the code.

If it's a brand new app I will design with the agent up front to make sure we're capturing everything and it is actually driving toward a specific outcome or else it can wander into 'productivity porn' territory where it's just creating reams of reports, plans, or 'evidence packs' and nothing real gets implemented.

I used to use a lot more sophistication in my workflow and skills and I still do use a few (impeccable is great for de-sloping design) but they've gotten a lot better at breaking down a plan and building without that.

I suspect we'll see (or what I want to see) is like an independent personal datahub and browser MCP which the assistant uses to conduct business as yourself so you have full visibility and control how it accesses your data or when it can use your cookies/credentials.

Also the dots/muse/etc browsers all are seemingly getting blocked so federating it out would seem to solve that problem too.

For B2B you really need to have face to face conversations that AI doesn't really help scale.

As others have noted her current prison time is for fraud against Walton, Ellison, Murdoch etc - the same family funds funneling money to the Trump-klepto-universe.

They are the ones who propped up the company without any independent testing and enabled the bad tests and are ruining our society with their dumb money.

This is family funds - Murdoch, Walton, Ellison. You think they shouldn't be responsible for doing due diligence? The device didn't work. They could've independently verified that at any point.

If there were users that were given bad tests, that was due to these investors not doing their jobs.

This is conflating users and investors. Investors have the obligation of diligence and it's clear here (and with FTX) they didn't. If they did, neither would be happened.

Also it's exactly the opposite, scamming the public is fine - United healthcare can scam the government out of billions, no one goes to jail, minor fees but it's +ROI for them.

Murdoch's lose an investment after YOLO-ing a few hundred mil and no diligence and they get the feds to give jail time?

I mean in do you really have a lot of sympathy for the oligarchic tax advantaged zombie family hedge funds that made a bet and lost?

They didn't do their diligence and do kinda deserve to lose their money.

I put in a bit more effort to de-slop and now it is looking a lot better plus I bought a domain: all-in-check.com

Thanks, trying a regen now..

This is honestly a bit AI slop but after checking a few claims manually I setup a bot to auto-review each ep. I'll publish the prompts in GH if this gets any traction, definitely interested in figuring out how to get better signal here and reduce the "AI voice".

I added a pivot ep just to demonstrate it's equally harsh and non-partisian in its grading.

Every other vendor has injected so many obtrusive ads and spyware Apple has won this product area by default.

The PR->merge->deploy SDLC is probably not what we want anymore.

Something like continuously delivery and validation through the whole stack so you can develop an incremental change, validate it as much as you can locally, then validate it as much as you can in the cloud using integration tests with synthetic, then with replayed production, and then finally allow it become production. If stuff starts breaking, you have your AI swam go in, identify the culprit, bounce it, and the train never stops.

I'm pretty excited by self hosted agents or maybe just self hosted MCPs to allow remote agents like Dots or Muse gated access to your data. Amazon blocks Muse/Dots but they can't block every DO or Cloudflare egress IP.

Excellent podcast, still not sure about their stack. No GPU support is a hard sell these days.

It's just fraud, but they just know the current government will not prosecute so it's 'legal'. If they do, eventually, get held accountable the fine will likely be less than the savings.

Sadly needed because insanely Google is allowed to show mitm/malicious ads when people search for government services.

Yeah this is other other-side of the coin. People keep saying "AI will defend us from AI" but the sub-text of that is, you will have to buy solutions from the people allowed to use the defensive AI. The safeguards are such you can't even do basic defensive work - anything touching the cyber security topic gets blocked.

It's pretty frustrating to do cyber security work and not have access to the best models. OAI is a little more liberal here and I was able to get access to daybreak-blue but I have to use the lesser last-generation models. Essentially this gives a small number of orgs a huge advantage in the 'application layer' for that domain (including OAI or ANT themselves).

Don't forget the unbounded liability that is the US healthcare system!

I think the problem is the parallel construction and lack of warrant

My hot take is that open models don't really save you money and introduce more router complexity and security risk (because you're now sending your company data through more less trustworthy providers). Look at cost per task not cost per token and the pareto curve is largely owned by closed models.

Just use Fable 5.1/Opus max for the hardest problems, GPT Sol high as your workhorse, and maybe terra for async batch stuff you don't really care about. Gemini 3.8 High also looks pretty good and is quite fast if you're already a GCP shop. You can basically benefit from open models without using them because they force the frontier models to be cheaper.

We should just progressively tax monopolies. Companies will break themselves up to compete, no decade long DOJ case needed.

Agreed

It's not just malicious software ads, they also sell ads to entities mitming every government service.

Seems like lead-generation for Vanta or other agentic-security tools? I tried to signup to take a look at the MCP and am forced to "Become a contributor" and supply a bunch of details to continue.

I like 1-2 days in-office if the team is actually local but the problem is that companies can't decide what they want (in-office vs cheaper LCOL labor) and it ends up being the worst of both worlds where you drive into the office just to sit alone in conference rooms on video calls with people in other offices.

I call it 'agent speak' and it's pretty exhausting. Another pattern is it'll emit dense enumerations of a possible answer and then emit dense enumerations of what isn't the answer. It's still helpful to help rough out/frame a document but then you have to go in and edit it pretty heavily. "Make it shorter", "Make it more concise" doesn't seem help all that much because it still keeps a lot of the same patterns.

Competition and Open models will keep AI providers more honest - Google and Amazon are monopolies so they can enshittify and users have no recourse.