siliconc0w

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.

I'm hoping AI will disintermediate most retail and we'll finally get an aggregator that isn't going to add a tax on both sides of the market.

"Buy me the best Air Fryer" -> finds the best given my preferences and constraints and directly purchases it from the cheapest distributor

Enterprises don't just send their data to any company with GPUs - they need security reviews, contacts, etc. They're assuredly using BYOK - not the marketplace.

I don't really understand the value -add, so far "smart" routing currently produces worse results at higher cost and static routing can be done yourself trivially. Plus the risk of sending the most sensitive data you have to a third party.

What I like to do is to go back and forth on the spec, break it into very detailed tasks and milestones, and then set a /goal to complete the milestone and verify. Each task is verified with 'fresh eyes' or a clear context.

That said, I'd really like to see data to compare which approach works the best.

That is more FOMO and maybe some end-game where SpaceX essentially owns a vertical slice of ISP+Datacenter+AI+application stack.

Competition keeps service quality high and pricing low - even if you aren't using Grok, the mere existence of Grok keeps pricing for whatever provider you use lower and service faster and more reliable.

There are cost externalities that generally unaccounted for- maintaining an underclass paid subsistence wages or using disposable products that as a society you eventually have to grapple with.

There is a tipping point between procedurally generating everything in SVG to maybe giving them tool access to something like 3dsmax (or having them build and then use a tool to do the thing vs doing the thing).

DSLs for AI doesn't really make sense- they were trained on existing graphics libraries and are pretty good at them.

Maybe a long term play would be putting this out and creating a "graphics bench" to entice the labs to overfit on your DSL but that seems like a lot of work

Built in replicated database and filesystem is pretty rad. Fly leaves this as an exercise for the user.

You should license the backend to enterprises that want a cheaper/lighter weight option to kubernetes for the surge of agentically coded internal apps.

I do wonder why not just use cheaper models like Luna or Grok instead of dealing with routing. DeepSWE has Grok at 54% for $2.42 or 67% for $3.03 for Luna. Tokenless Pro is 6.67 for the same performance as Luna.

I don't think it changes much for good managers. It should always be able setting people and processes up so the team can land durable measurable impact. The managers that thought the job of software engineers was to write code were bad managers. PRs or LoC were never good metrics.

A few things:

0) big companies already are very comfortable using contracts to trust other people with their data. Maybe if they're inflexible on the 30-day requirement for fable some orgs will opt out but by-and-large it's already happening and it's not a blocker.

1) Cost will be a blocker. The level of token spend is untenable and the pareto curve is flattening. Most orgs are going to default to either using a distilled model from China or a distilled model from the model companies (e.g, Sonnet 5). It'd behoove Claude/OpenAI to offer a model router before another vendor wins that area.

2) Karp is selling his book. No one knows or cares what an 'ontology' is. From what I can tell, company's product is a tool that helps governments bomb people.

Seems like rationalization - PE firms establish regional monopolies and jack up rates. They do it with every type of service. Just accept that we're all serfs under the billionaires who extract rents over every aspect of our lives.

Google pays their oncall a % of their full-time base salary depending on the oncall tier (5 min response time vs 30 minutes).

This should probably be required - there is a different mindset and set of restrictions when you're expected to pick up a page. It also forces companies to use on-call judiciously - not every service needs a 5 min SLO.

What are the odds this is going to become another avenue for grift - magically any companies the trump family invests in are going to get access. Any companies that aren't sufficiently 'loyal' to the regime will have to wait or may never get access.

The reality is that there is no escaping management. A high level IC is pretty much just an unofficial manager. You are responsible for a large technical area - it's your job to meet with stakeholders, design the roadmap, build durable team processes to maintain velocity, mentor and identify the right work for the right people. You may not have back to back meetings but between 1:1s, stakeholders, projects, and ad-hoc fires they're likely still a majority your schedule. You are expected to lead without authority. Leadership will change priorities and reorganize the teams every six months. If you're focused and deliberate, you can maybe get can your project landed before the next reorg.