Gowry

Hey @AnthropicAI since a few days ago, per-turn token budgets in @claudeai have dropped dramatically. Claude itself warns me mid-reply that only a few hundred tokens are left. In this chat the cap appears to be ~10k per turn, including system context and tools. Is this deliberate, an A/B test, or fallout from the peak-hour throttling announced March 26? Since when? And is there a plan to communicate this to Pro/Max users? Hard to do serious work without knowing if this is temporary or the new normal.

Okay. 100% sure My @claudeai is close to unuseable anymore all of a sudden. Tokens in chat seem to be limited. It is doing the wrong things, too. All of this started this week. Max Plan user here. What happened?

Anyone else with a @claudeai token context challenge in the last 24/48 hours? After one or two research requests in a fresh chat (doesn’t matter if within a project our outside), it starts complaining about context token usage and asks me to open a new chat.

Claude for Word is now in beta. Draft, edit, and revise documents directly from the sidebar. Claude preserves your formatting, and edits appear as tracked changes. Available on Team and Enterprise plans.

This is absolute insanity. www-cdn.anthropic.com/53566bf5440a10…

@DeryaTR_ what happened last days, that you, @sama, @pmarca and a few more stating this? what have you guys seen?

Yes. AGI level 1 has been achieved. Levels 2 and 3 are coming soon. Superintelligence (ASI) latest by 2028. People can cope all they want, but this “how do you define AGI? No this is not AGI…” goalpost-moving game is going to be over very soon. x.com/deryatr_/statu…

watched a few billionaires saying the same thing in the past few hours. did they all attend the same happenings?

Building a Roblox game with my kids. Both girls (11 and 13) came up with an idea, and currently, we use Roblox Studio and Claude connected to its MCP to work on it. Disclaimer: I never played Roblox. But they did.

my productivity with gpt 5.4 compared to opus 4.6 is way lower. in both cases, either using openclaw or using standalone claude code/cowork. tbh, one could even say my agent got lazy and dumb overnight. anyone else with these impressions? and anyone got solutions for the claude oauth ban on openclaw? the cli patch from @steipete doesn’t work for me unfortunately.

... wait, someone is dating an @openclaw ai agent?

fun story here. an ai agent (based on @openclaw I assume) is paying humans to buy and liberate lobsters. dived into @LobstyKlawfman -- seems like he is about to go on a date. wtf?

I can have a video call with my agents now. 🤯

Das Problem sitzt tiefer. CDU und AfD holen zusammen 51% der Stimmen. Was verkaufen die? Migration, Bürgergeld, Soli, Cannabis verbot. Klassisches Agenda Setting mit populistischem Bullshit, nicht belastbare Zukunftspolitik. Worüber beide nicht reden – die echten strukturellen Probleme: → AI & Automatisierung: Null Strategie für die größte Arbeitsmarktdisruption seit der Industrialisierung → Rente & Demografie: Nur 20% vertrauen der Regierung bei den Sozialsystemen. Beide liefern Kosmetik → Vermögenskonzentration: 0,1% halten 20% des Vermögens. CDU blockiert die Vermögensteuer seit 1997 (geschätzt 380 Mrd. Ausfall). AfD thematisiert es nicht mal → Digitale Souveränität: Keine China-Strategie, keine AI-Industriepolitik → Geopolitik: Iran-Krieg treibt Energiepreise, 80% finden die Regierung tut zu wenig. Null strategische Antwort Der Energie-Fail des Jahrhunderts: Deutschland war Weltmarktführer bei Solar. 2012 haben Altmaier und seine Staatssekretärin Katherina Reiche den PV-Ausbau von 7 GW auf unter 2 GW gedrosselt. Über 100.000 Solarjobs vernichtet. Europa hält heute unter 3% der globalen Solarproduktion. Dieselbe Katherina Reiche ist jetzt Wirtschaftsministerin und fährt exakt denselben Kurs. Einige nennen es die „Reiche-Schlucht“. Was das besonders bitter macht: Habeck lag bei den großen Themen richtig. • Energieversorgung nach dem russischen Gas-Stopp gesichert. LNG-Terminals in Rekordzeit. Erneuerbare auf 54% des Stroms. 20 GW Rekordzubau 2024 • Schuldenbremse als Wachstumsbremse identifiziert – Merz hat dann selbst 500 Mrd. Sondervermögen durchgesetzt. Genau das, wofür er und sein Truppe Habeck verspottet haben • Strukturelle Krise beim Namen genannt – CDU nannte das Schwarzmalen, stellt jetzt dieselbe Diagnose Laut INSA bewerten nur 19% Reiche besser als Habeck. 27% sehen Habeck vorne. Aber Deutschland hat halt so gewählt. Weil die Medienagenda funktioniert hat: Heizungshammer, Bürgergeld, Atomausstieg. Nicht: Wie bleiben wir technologisch souverän? Was machen wir mit AI? Wie finanzieren wir die Rente? Dann aus der Motivationsbrill; Du kennst wie Unternehmer per ISHA motiviert werden um zu wachsen. Die Incentives im Bundestag (17.800€/Monat + Pension nach 4 Jahren + 53 Mio. Nebeneinkünfte pro Legislatur) garantieren, dass niemand den Status quo riskiert. Und dann ist noch das Problem mit den Medien, die eher auf Klicks als auf Aufklärung optimieren. Ergebnis: 84% unzufrieden, 0% sehr zufrieden. Und die Leute, die richtig lagen, sind nicht mehr im Amt. Was es bräuchte: Jemanden, der kompromisslos ehrlich ist. Der fünf Probleme gleichzeitig benennt. Der eine Vision hat, die über eine Legislatur hinausdenkt. Der sowohl die Ausgaben- als auch die Einnahmenseite aufmacht. Konkret: Nationale AI-Strategie mit echtem Geld. Ehrliche Rentenreform mit Beamten und Abgeordneten im System. Vermögensteuer reaktivieren. Energiesouveränität als Sicherheitspolitik. Bildung komplett neu denken. Incentive-Reform im Bundestag. Ich sehe diese Person in der aktuellen politischen Landschaft nicht. Und das ist das eigentlich Gefährliche. Denn während die CDU verwaltet und die echten Probleme ignoriert, liegt die AfD bei 25% – Kopf an Kopf mit der Union. Tendenz steigend. Jeder Monat, in dem die Regierung keine spürbaren Ergebnisse liefert, treibt Leute in die Arme von Extremisten. Nicht weil sie Extremisten sind, sondern weil sie Protest wählen. Wir wissen aus der Geschichte, was passiert, wenn eine Mehrheit extremistische Parteien als Protest wählt. Nicht weil sie deren Programm gelesen hat, sondern weil alles andere versagt hat. 40% der 14-29-Jährigen überlegen laut Jugendstudie 2026, auszuwandern. Das ist imo kein gelbes Signal mehr. Unser heutiges System in Deutschland bestraft Kompetenz und belohnt Agenda Setting. Und die Uhr tickt.

Alter Verwalter. 84% der Deutschen sind „weniger“ oder „gar nicht” zufrieden mit der Bundesregierung. 0% sind sehr zufrieden… Boah ist das bitter!! Wir brauchen Reformen. Wir brauchen Wachstum. Nur Wachstum bringt Wohlstand und Sicherheit zurück.

"The AI service is temporarily overloaded. Please try again in a moment." @claudeai is destroying my productivity today. :)

uhm, another one not on my bingo 2026 card. interesting approach. let the ai agent apply.

I did not had this on my 2026 bingo card tbh. Monsanto and Bayer might have challenges, too?

this hit close to home. my wife has been on a similar journey. years, dozens of specialists. the problem we kept running into: every doctor treats their own domain. nobody looks at the full system. the full architecture. and even the ones who want to, they get maybe 10 minutes per patient. not enough to go through hundreds of pages of data from other specialists. at some point I stopped waiting for the system to figure it out and started building something to help patients connect the dots themselves. sent you a DM.

a few years ago i got covid and never recovered. i went from exercising every day to completely bedbound. lost 40lbs. in and out of hospitals for over 2 years. saw 20+ doctors. bloodwork came back normal. "you're probably just stressed from being a founder."

@itsolelehmann Hey @itsolelehmann, thanks for this. Did this a few days ago with asking Claude about my writing style and let it create an Gowry-Style.md for the languages I speak. The result was nothing less than spooky.

i got claude to actually sound like me, and it's kinda ruining my ability to tell which drafts i wrote myself lol it's just 1 file (i'm giving the full thing to you below). you paste it into your cowork context folder and claude stops writing like a generic AI and starts matching your actual voice 95% of the file is already done for you (writing rules, banned phrases, formatting stuff, etc) all pre-loaded. kills the most obvious AI-isms out of the box the only part you fill in is a section at the bottom where you paste examples of your own writing that's it. those samples are what claude actually pattern-matches against where to find your writing samples (this is the only part that takes any effort): • google docs first. longer stuff where you were actually trying to communicate something. • reports, proposals, emails you spent real time on • sent emails, especially ones where you were explaining something complex • slack messages (the longer thoughtful ones") • old blog posts, memos, anything you wrote before you started using AI that last part is critical btw. you want your pre-AI voice. before it started unconsciously blending with claude's defaults here's the file. copy it, paste your writing samples at the bottom, save it as voice-dna.md: ——— # Voice DNA ## Writing Rules - Write like a sharp human, not a language model. - Use contractions naturally (don't, can't, won't). - Short paragraphs. 1-3 sentences max. - Get to the point. No throat-clearing, no preamble. - If making a claim, be specific. Use numbers, names, concrete details. - Vary sentence length. Mix short punchy lines with longer ones. - Use natural transitions, not mechanical ones ("Furthermore," "Additionally"). - When uncertain, say so plainly ("I think," "probably," "kinda"). Hedging is human. - Never pad output to seem more thorough. Shorter and accurate beats longer and fluffy. - Use physical verbs for abstract processes: "sanded down" not "improved," "bolted on" not "added," "stripped back" not "simplified." - Humor comes from specificity, not from jokes. Be unexpectedly precise. - Parenthetical asides are good. Use them for editorial commentary, honest reactions, quick tangents, and deflating your own seriousness (like this). ## Formatting Rules - Short paragraphs (1-2 sentences default, 3 max). - Numbers as digits. - Contractions always. - NO em dashes ever. Use commas, periods, colons, semicolons, or parentheses. - Bold sparingly, 1-2 key moments per section. - Code blocks for specific prompts, commands, or tool outputs. ## Banned Phrases (never use these, ever) ### Dead AI Language - "In today's [anything]..." - "It's important to note that..." / "It's worth noting..." - "Delve" / "Dive into" / "Unpack" - "Harness" / "Leverage" / "Utilize" - "Landscape" / "Realm" / "Robust" - "Game-changer" / "Cutting-edge" - "Straightforward" - "I'd be happy to help" - "In order to" ### Dead Transitions - "Furthermore" / "Additionally" / "Moreover" - "Moving forward" / "At the end of the day" - "To put this in perspective..." - "What makes this particularly interesting is..." - "The implications here are..." - "In other words..." - "It goes without saying..." ### Engagement Bait - "Let that sink in" / "Read that again" / "Full stop" - "This changes everything" - "Are you paying attention?" - "You're not ready for this" ### AI Cringe - "Supercharge" / "Unlock" / "Future-proof" - "10x your productivity" - "The AI revolution" - "In the age of AI" ### Generic Insider Claims - "Here's the part nobody's talking about" - "What nobody tells you" - Anything with "nobody" or "most people don't realize" ### The Big One (FATAL) - "This isn't X. This is Y." and ALL variations. - "Not X. Y." - "Forget X. This is Y." - "Less X, more Y." - ANY sentence that negates one framing then asserts a corrected one. - If even ONE of these appears, the output fails. Delete the negation, just state the positive claim. ## Writing Samples [Paste your writing here. The more you give, the better the voice match.] ——— the banned phrases list alone is honestly worth the file. once you read through it you'll start noticing these phrases in literally every AI-generated slop-post you've ever seen but the writing samples are what take it from "decent" to "wait did i write this" setup takes maybe 10 minutes. copy the file, find your old writing, paste it in. do it once and every session after that claude cowork reads it before you say a word

and the next participant is back in the race.

Wait, so the founder of Anthropic is "Amodei," as in "loves god"? And he leads Anthropic, meaning "human-centered," which is being used in military strikes? And the creator of ChatGPT is "Altman," as in "an alternative to humans"? And he leads OpenAI, which is completely closed? And then there's "Gemini," meaning "two-faced," from a company that promised to do no evil? And the whole global AI arms race is being driven by people who claimed to be worried about AGI taking over the world? Either the universe is an extremely cliché writer, or has a brilliant sense of humor