Austin

Another day of blazing fast @TMobile Home Internet.🙄I'll reboot the gateway, and it'll jump to 300 mbps. So I'm just getting cheated all day? I've never had to babysit my internet so hard.

@HedgieMarkets Feels like you're ignoring the impact of local models. Cloud AI price hikes are capped somewhere based on local options. Plus, integration/infrastructure moat is less valuable thanks to open source options. To think that these companies have unlimited pricing power is laughable.

🦔Microsoft canceled its internal Claude Code licenses this week after token-based billing made the cost untenable, even for a company with effectively infinite cloud resources. Uber's CTO sent an internal memo warning the company burned through its entire 2026 AI budget in just four months. American AI software prices have jumped 20% to 37%, and GitHub (owned by Microsoft) is dropping flat-rate plans for usage-based billing across its products. My Take The AI subsidy era is ending in real time. The same company that put $13 billion into OpenAI and built the Azure infrastructure powering most of Anthropic's compute just looked at the bill from a competitor's coding tool and decided it was not worth paying. That is not a productivity failure on Anthropic's end. Token-based pricing is forcing every enterprise customer to confront the actual cost of running these models at scale, and the number turns out to be far higher than the flat-rate experiments suggested. This ties directly to my Gemini Flash post yesterday. Anthropic, OpenAI, and Google all raised effective prices in the last six months. Enterprises that built workflows assuming AI costs would keep falling are now watching annual budgets evaporate in months. Two outcomes look likely from here. Either enterprises scale back AI usage to fit budgets, which slows the revenue ramp the labs need to justify their valuations ahead of IPOs, or the labs cut prices and absorb the losses, which makes the unit economics worse at exactly the wrong moment. Both paths land in the same place, the numbers stop working, and somebody has to take the writedown. Hedgie🤗

I don't know much about this brand, but this seems like very bad taste. Terrible optics. @LiquidDeath

@RonDeSantis About 18% in taxes and fees for EV charging. Criminal already. Plus the energy costs are already marked up.

Right, it’s inaccurate to claim EV owners “don’t pay tax“ — they pay various taxes when utilizing a charger depending on the jurisdiction.

@kylejeong Are you kidding? This is amazing... Can't wait to try this tomorrow when my credits refresh after being burned from learning a few sites the hard way today. Perfect timing.

.@Google: "Building the world's most capable Al assistant." The AI Assistant:

My personal experience with EV chargers: @tesla Superchargers: <2% failure rate. @insideFPL EVolution chargers: ~40% failure rate. @ChargePointnet chargers: ~80% failure rate. Invest your infrastructure dollars accordingly.

@HermesAgentTips I'd be happy with an @nvidia DGX Spark.

Local LLM people If money wasn’t an issue, what hardware would you be running right now?

@NVIDIA_AI_PC @NousResearch @ollama 🤤🤤🤤💸💸💸

Run @NousResearch's Hermes Agent fully locally on DGX Spark. 🚀 Our newest playbook shows you how to get set up via @Ollama step by step. 👇

Almost a month later and I am STILL impressed with @ollama Pro. BUT, I would like to see: 1) Cloud usage by model (to see how tiered modeling improves my credit burn). 2) More specific reset timing. "Resets in 2 days" is too ambiguous. What date? What time?! 🤓

Been using @ollama Pro in Hermes (GLM 5.1 cloud) for a few days and I’m legitimately impressed. Very generous usage, rock solid performance. It was my backup plan for the now-rapidly-depleting Codex oAuth, but my limits reset yesterday and I still haven’t switched back to Codex.

Standbye-bye Sorry, @elonmusk and @Starlink, raising the price of standby 50%, for a service that is OFF, and just-in-case has snapped my price elasticity. (And sending the email at 8:32am so it's the first thing I see on Sat AM? Nice start to the day!)

@TMobileHelp Thanks, AlexisG, but I've talked to too many Tmo tech agents, replaced the gateway and nothing helps. I'm giving up and moving on. Here's the issue in a snapshot. Trickle internet unless I restart the gateway often.

@omicAustin I am confident I can help fix the issues you’re experiencing with service, Austin. Please send me a DM with details about what is going on and I will work fast to get this figured out. ^AlexisG

If anyone is wondering, @TMobile Home Internet is the most unreliable Internet I've ever paid for. I regret switching.

@AlexFinn Congratulations on the success. Well deserved...

Unreal. Truly has been the greatest year of my life 200k subs on YT. Soon will hit 500k on X When I was a kid I was supremely unconfident and miserable I failed almost every class I took and had basically no friends It wasn't until I hit 30 that I found my passion + happiness If you would have told me at 22 I was going to have my own business, would be teaching millions of people on the internet about technology every day, and would have found true happiness, I'd think you were a lunatic I didn't think happiness or enjoying my career was ever in the cards for me. Luckily I was wrong To everyone who has supported me every step of the way, thank you. Truly. Love you all and can't wait for what's next

@hnshah @oliverhenry One to piss you off and one to be productive? 🤓 I'm also curious about the different purposes.

@oliverhenry I use both. They serve different purposes.

I have still yet to meet someone actually using hermes. Whenever I mention it, i just get bots in my replies. Openclaw 4 lyf

@oliverhenry I moved from OC about a month ago. Not looking back. Been very impressed. I thought the Hermes hype was BS too. OC had amazing promise, Hermes actually works for me.

I wish someone would make this...

@OpenAI We've reached the phase of using badass music in AI model demos. I'm down.

Introducing GPT-5.5 A new class of intelligence for real work and powering agents, built to understand complex goals, use tools, check its work, and carry more tasks through to completion. It marks a new way of getting computer work done. Now available in ChatGPT and Codex.

JMO, but this seems like poor planning @DOGE_IRS