Exploring the frontier // Nomad since '19
Joined July 2017
- Tweets4.1K
- Following1.8K
- Followers7K
- Likes10.7K
Pinned Tweet
Anthropic just pulled the oldest trick in SaaS pricing.
I pay $200/mo for Claude Max. My limits have been noticeably worse this past week.
Now they announce 2x off-peak usage for two weeks. Sounds generous.
But here’s what actually happens: limits quietly drop, a temporary 2x makes the reduced limit feel normal, the promo ends, and you’re left at a baseline lower than where you started. You just didn’t notice the downgrade because the 2x absorbed the transition.
These AI plans are massively subsidized. The raw compute behind a heavy user costs multiples of the subscription price. Every move like this is the subsidy quietly correcting.
Very sneaky, Anthropic.
OpenAI clearly rushed Dots just so competitors wouldn't run away with the category
After using it for a couple days, the product is clearly premature. Give it a few simple tasks and it starts tripping over itself and becomes useless
Probably the worst OpenAI release in recent history
Anyone that’s actually creating designs knows that Opus is far superior to Astra
It’s not even close
BREAKING: GPT-6 Astra by @OpenAI has taken the #1 spot across 4 of Design Arena’s leaderboards: 3D Design (1484), Frontend (1397), Full Stack (1355), and Image-to-HTML (1272).
This is an impressive takeover since its release just a month ago - in particular, it excels at 3D Design (see example below):
Called this in June when @semianalysis_ broke down the numbers
The $200 plans were running on a 40-70x compute subsidy. That was never going to survive as a flat rate. The correction was always going to come as usage cuts
OpenAI is now cutting the $200 Codex plan to half the API spend allocation.
The beginning of the end is here. Subscriptions and metered API pricing are converging
Rent is due
xitter.cf/thsottiaux/status/2104…
Hi,
Tomorrow we are re-opening the Pro $200 subscriptions to new subscribers, but together with it we are also changing how we calculate the usage for it. In effect, if you do the math, it will net out at half the dollar in API spend compared to the old Pro $200 plan.
Now that it's said, let me explain why this is happening and why you will still get more work done than if you were on the Pro $200 subscription one month ago.
(a) We didn't want to compromise in other ways and are committing to not reintroducing the 5h limit, so that you can fully use the weekly usage when you want.
(b) On the subscription, we guarantee that over time you always get more work done and with an increasing level of quality. This means that you will continue to get more value per dollar spent as a result of models getting more efficient and us passing down the improvements in the form of API price reductions.
(c) We don't want to put an incentive on ourselves to artificially inflate the API list prices to make it look like you are getting a lot (and workaround it through discounts, etc). Instead we want to continue to both rapidly reduce prices and increase capabilities of models on the API. This week we introduced GPT-6 Sol and GPT-6 Luna at 50% of their previous price. Over time, we see prices go low enough that it makes sense for most to buy usage as needed without there being a significant gap between what you get in a subscription and what you get in the API for a dollar spent.
(d) Tomorrow, we are adding more things to the subscription that won't draw on the usage, I won't reveal what that is yet.
I wanted to be transparent before all the big announcements tomorrow. Lots of new exciting things are coming to the subscriptions that will make it super compelling, but I wanted to make sure to share this change ahead of time so you can all understand it before we shower you with good news.
Codexingly,
Tibo
Inspired by @Plasma
Porsche buyers, this one's for you.
20% cashback on Porsche. Forever.
Readers added context they thought people might want to know
Cashback is capped at $10 per user.
Since the cheapest Porsche currently available is the Macan starting at $65,400, the effective cashback would be 0.015%.
x.com/aadvark89/stat…
porsche.com/usa/models/mac…
When does the rest of the world get Muse @alexandr_wang @finkd ?
A lot of builders internationally, including myself, want to build Muse connectors. We want to test Muse and see what integrations are actually worth building, but it is impossible to build connectors for an agent you are geoblocked from running
xitter.cf/finkd/status/210108467…
Opening access for developers to build Muse connectors. You bring the API -- Muse brings the agent, the browser, and the context of what the person actually wants. People reach your service just by asking for it, and their agent takes it from there.
New connectors are live today. Come build with us. muse.ai/platform
Typing on glass feels genuinely prehistoric at this point
It is hard enough typing on a computer, but once you use voice dictation for everything, having to tap out a sentence or two on mobile feels completely broken
Voice over everything
Current models are already smart enough to handle 95% of what most people need. Smarter frontier models will matter for scientific discovery and edge cases, but the raw intelligence gains feel incremental right now
The real step function is computer use:
Chatbots → coding agents → knowledge work → computer use
Models operating directly on your screen is the unlock that actually changes who can use this stuff
Apps like Grokbot, Instinct, and Muse feel different because the agent is driving the interface itself. That's how you get someone like my mom running autonomous workflows instead of copy-pasting into a chat window
Demand for AI still has no visible ceiling and the trade is very alive, but the next wave will be driven by software that knows how to drive software