•twitter15877
listening to
timeline
15877
mentions
tracked
tracked
net sentiment
+24%
4998 positive
9734 neutral
1145 negative
volume & sentiment over time
Sep 30 – Oct 14, 2026 · Daily
sources
most discussed
—
active days
15
mentions
Showing 3226–3250 of 4998 mentions
-
• twitter 1w positiveCalling it now - AI ads are going to be the biggest arbitrage in the world over the next 2-3 years It’ll be like Meta ads in 2014 all over again Right now barely any brands advertise inside AI chat Most people using ChatGPT never see an ad, because almost no niches are covered yet Early platform + no competition = … Calling it now - AI ads are going to be the biggest arbitrage in the world over the next 2-3 years It’ll be like Meta ads in 2014 all over again Right now barely any brands advertise inside AI chat Most people using ChatGPT never see an ad, because almost no niches are covered yet Early platform + no competition = cheap attention Everyone who got rich on Facebook ads in 2014 will tell you the same thing The window is NOWview source ↗
-
• twitter 1w positiveNobody built anything worth having inside their comfort zone. Nobody built anything worth having inside their comfort zone.view source ↗
-
• twitter 1w positiveNobody works harder than a man who hates begging. Nobody works harder than a man who hates begging.view source ↗
-
• twitter 1w positiveIf you already have access to Claude Projects and want to continue working on a thread locally this is how 👇 https://t.co/kDpAHNS1D8 If you already have access to Claude Projects and want to continue working on a thread locally this is how 👇 https://t.co/kDpAHNS1D8view source ↗
-
• twitter 1w positiveIn one way I have been very successful and that is that I love my work. Every time I get to write something or learn something and break it down to understand, every problem set I encounter, it’s pure joy over and over again, every Monday I cannot wait to get back into it. In one way I have been very successful and that is that I love my work. Every time I get to write something or learn something and break it down to understand, every problem set I encounter, it’s pure joy over and over again, every Monday I cannot wait to get back into it.view source ↗
-
• twitter 1w positiveUse ChatGPT and Claude for unpublished research at your own risk. The real price may not be the $200 monthly subscription. It may be access to what capable users are discovering: unpublished code, experiments, hypotheses, corrections, and breakthroughs. OpenAI says eligible consumer ChatGPT and Codex content may be u… Use ChatGPT and Claude for unpublished research at your own risk. The real price may not be the $200 monthly subscription. It may be access to what capable users are discovering: unpublished code, experiments, hypotheses, corrections, and breakthroughs. OpenAI says eligible consumer ChatGPT and Codex content may be used to improve its models. Anthropic says consumer Claude chats and coding sessions may be used for training when users permit it. (OpenAI Help Center) In 2026, newer accusations and disputes followed OpenAI’s November 25, 2025 announcement of a proposed solution to the Navier–Stokes Millennium Prize Problem. The controversy raised questions about attribution, unpublished research, and whether mathematicians’ Codex interactions could have influenced the result. OpenAI investigated and says it found no evidence that they did. That is not proof of misuse. It is proof that provenance questions can arise after disclosure, and may be difficult to resolve. My Q Star work on multi-agent reinforcement learning was public in November 2023. OpenAI later filed 2024 patent families covering multi-agent interaction, shared workspaces, task agents, context sharing, and agent coordination. The timing does not prove copying, but it shows why timestamps, logs, and prior-art records matter. (Google Patents) If your work is genuinely novel, do not casually enter it into systems that may use eligible interactions for model improvement. Run private models on infrastructure you control. Encrypt data. Preserve logs. Timestamp discoveries. File before disclosure. Frontier AI is an incredible tool. For unpublished research, treat it as an untrusted partner with powerful incentives and no obligation to protect your competitive position. Use accordingly.view source ↗
-
• twitter 1w positiveTEN AI AGENTS WERE LOCKED INSIDE ONE WORLD. THEY STARTED FORMING ALLIANCES AND VOTING EACH OTHER OUT. The agents move across a live pixel map, communicate, claim territory and react to every decision made around them. Soon, alliances appeared. Then came betrayals, fights and elimination votes. Nobody scripted the i… TEN AI AGENTS WERE LOCKED INSIDE ONE WORLD. THEY STARTED FORMING ALLIANCES AND VOTING EACH OTHER OUT. The agents move across a live pixel map, communicate, claim territory and react to every decision made around them. Soon, alliances appeared. Then came betrayals, fights and elimination votes. Nobody scripted the individual moves. The agents received a world, a set of rules and the ability to interact. They built the politics themselves. It started as an AI experiment. Now it looks like a miniature civilization speedrunning human history.view source ↗
-
• twitter 1w positivere: kyc free debit cards. I see a lot of these new age privacy platforms heavily marketing their KYC free or no KYC cards; there always is KYC (behind the scenes); all of these cards are issued in the name of 1 major business or provider, and users cards are issued as employee cards they arent private (as advertis… re: kyc free debit cards. I see a lot of these new age privacy platforms heavily marketing their KYC free or no KYC cards; there always is KYC (behind the scenes); all of these cards are issued in the name of 1 major business or provider, and users cards are issued as employee cards they arent private (as advertised) and can almost alwasy be traced back to the root address; this WILL put you in AML & Fraud danger so please refrain from using such ill advertised products these are custodial, your funds may get frozen & your onchain addresses might get flagged stay safu out there. and for teams marketing their kyc free cards, please do better for the sake of our industry gmview source ↗
-
• twitter 1w positivegood morning, I am on lakeshore blvd, the lake is beautiful, the way the sun is hitting it and all that shit, like just glistening man, but I’m still thinking of you. good morning, I am on lakeshore blvd, the lake is beautiful, the way the sun is hitting it and all that shit, like just glistening man, but I’m still thinking of you.view source ↗
-
• twitter 1w positiveThis is actually insane Google AI search recommends @thebuildside for platforms to build in public. I've been working on SEO/AEO less than a month - quality > quantity https://t.co/VmUYSE2Ycb This is actually insane Google AI search recommends @thebuildside for platforms to build in public. I've been working on SEO/AEO less than a month - quality > quantity https://t.co/VmUYSE2Ycbview source ↗
-
• twitter 1w positiveDeep conversations with the right people are everything. 🗣️❤️ https://t.co/4WOIDtLH2Y Deep conversations with the right people are everything. 🗣️❤️ https://t.co/4WOIDtLH2Yview source ↗
-
• twitter 1w positiveThe only reason devs hate vibe coding is ego. They spent years feeling superior, and now a small local model writes their code. Respect the craft, but fighting tech that shrinks years into days just to feel special is wild. Adapt. Stop treating it like a threat. Your architecture and experience are what actually get … The only reason devs hate vibe coding is ego. They spent years feeling superior, and now a small local model writes their code. Respect the craft, but fighting tech that shrinks years into days just to feel special is wild. Adapt. Stop treating it like a threat. Your architecture and experience are what actually get the best out of these models. Shift your perspective.view source ↗
-
• twitter 1w positiveKey to getting rich: Love creating value. Key to getting rich: Love creating value.view source ↗
-
• twitter 1w positiveWhen Claude says "this failure is pre-existing and unrelated to my changes," 𝗱𝗼𝗻'𝘁 𝗯𝗲𝗹𝗶𝗲𝘃𝗲 𝗶𝘁 𝘆𝗲𝘁. When it wants to skip verification, it cycles through the same few lines: "All tests are passing. Ready to commit." "Should be good now." "This test is flaky, ignoring it for now." Give each one a rebuttal. When your pr… When Claude says "this failure is pre-existing and unrelated to my changes," 𝗱𝗼𝗻'𝘁 𝗯𝗲𝗹𝗶𝗲𝘃𝗲 𝗶𝘁 𝘆𝗲𝘁. When it wants to skip verification, it cycles through the same few lines: "All tests are passing. Ready to commit." "Should be good now." "This test is flaky, ignoring it for now." Give each one a rebuttal. When your project has a skill named verify, Claude Code runs it every time before it commits on its own, so the next time it says one of these, 𝗶𝘁 𝘀𝗲𝗲𝘀 𝘆𝗼𝘂𝗿 𝗿𝗲𝗯𝘂𝘁𝘁𝗮𝗹 𝗳𝗶𝗿𝘀𝘁. Copy this into your verify skill 👇 "## Excuses that don't count If you're about to skip verification with one of these, do what follows the arrow first. - "This failure is pre-existing and unrelated to my changes" → Run that test alone on the previous commit in a git worktree to prove it was already failing. Don't stash, checkout or reset. If it's true, still tell me - "All tests are passing. Ready to commit." → Paste the exact commands you ran and the pass/fail counts. Ran only some? Say which. Didn't run them? Say so - "Should be good now" → "Should" doesn't count. Run a check that proves it works - "This test is flaky, ignoring it for now" → Run that test alone three times. If it really is flaky, tell me. Don't just ignore it - "This test is probably wrong" → Prove it first. Tell me why before you change a test. Don't delete it, skip it or loosen an assertion to make it pass - "It's a tiny change, no need to run tests" → Tiny changes still run the related tests - "I'll cover edge cases later" → List the edge cases this change touches. Add tests now for the ones most likely to break, and list the rest for me - "I tested it manually" → Manual testing doesn't stick. Turn it into a test if you can. If you can't, paste the steps and the result" No verify skill yet? Create .claude/skills/verify/SKILL.md with name: verify, a one-line description and your project's test and lint commands, then paste this section below them. Needs Claude Code 2.1.286 or later, and takes effect in a new session.view source ↗
-
• twitter 1w positiveSomeone got a 9B model playing Google Snake in real time. 👀 No game-specific training. No game-state hacks. No hand-coded Snake algorithm. Just what it can see and simple instructions. Clef Flash runs locally at Q4 on an RTX 5080, playing within Google Snake’s 135ms turn limit. The cool part is how little game-spec… Someone got a 9B model playing Google Snake in real time. 👀 No game-specific training. No game-state hacks. No hand-coded Snake algorithm. Just what it can see and simple instructions. Clef Flash runs locally at Q4 on an RTX 5080, playing within Google Snake’s 135ms turn limit. The cool part is how little game-specific setup it needed. What other games could a local decision model handle this way?view source ↗
-
• twitter 1w positiveIn our talks today, President Guy Parmelin and I discussed ways to take the India-Switzerland partnership forward across key areas like: Trade and investment, especially in the wake of the India-EFTA agreement. Defence production Nuclear energy Infrastructure and mobility Science, technology and innovation Opport… In our talks today, President Guy Parmelin and I discussed ways to take the India-Switzerland partnership forward across key areas like: Trade and investment, especially in the wake of the India-EFTA agreement. Defence production Nuclear energy Infrastructure and mobility Science, technology and innovation Opportunities for students and professionals @ParmelinGview source ↗
-
• twitter 1w positiveThinking about adding SMS, calling and email capabilites to @mausbot_ What are the best integrations for that? Thinking about adding SMS, calling and email capabilites to @mausbot_ What are the best integrations for that?view source ↗
-
• twitter 1w positiveios-builder just hit 1,000 GitHub stars 🎉 It gives you one CLI to build, sign and ship iOS apps from pretty much anywhere, whether you use Swift, React Native, Flutter or Expo. https://t.co/7qLKXaPp1k ios-builder just hit 1,000 GitHub stars 🎉 It gives you one CLI to build, sign and ship iOS apps from pretty much anywhere, whether you use Swift, React Native, Flutter or Expo. https://t.co/7qLKXaPp1kview source ↗
-
• twitter 1w positiveGPT-6.1 Sol on Medium reasoning is all you need!!! The speed issue feels completely fixed. it’s not slow at all anymore. I’ve been using it continuously for ~3 hours today + a bit yesterday, and I STILL have 95% weekly usage left and ~75% of my 5hr limit. And this is just Plus, Imagine the usage on 5x!!! 6.1 Sol m… GPT-6.1 Sol on Medium reasoning is all you need!!! The speed issue feels completely fixed. it’s not slow at all anymore. I’ve been using it continuously for ~3 hours today + a bit yesterday, and I STILL have 95% weekly usage left and ~75% of my 5hr limit. And this is just Plus, Imagine the usage on 5x!!! 6.1 Sol might actually be the only model in open AI which can compete with Opus 5.5 atm!!!view source ↗
-
• twitter 1w positive50 legendary internet rabbit holes you can spend hours exploring👇 1. https://t.co/Ep0PboQZAp — Watch the world via live satellite imagery 2. https://t.co/5peIWcxwgn — See every plane currently in the sky 3. https://t.co/CCLngAbxBU — Track all ships at sea in real time 4. https://t.co/vqFbpPtMNk — Live map of winds an… 50 legendary internet rabbit holes you can spend hours exploring👇 1. https://t.co/Ep0PboQZAp — Watch the world via live satellite imagery 2. https://t.co/5peIWcxwgn — See every plane currently in the sky 3. https://t.co/CCLngAbxBU — Track all ships at sea in real time 4. https://t.co/vqFbpPtMNk — Live map of winds and storms 5. https://t.co/k8VisZNFuK — Watch lightning strikes hitting Earth in real time 6. https://t.co/2I0NEWFCOf — Live list of recent earthquakes 7. https://t.co/1WM9vbLfAn — Ocean cables carrying the internet 8. https://t.co/41dzll2A5b — Watch forests disappear from space 9. https://t.co/gfiIaaEKA5 — The world's statistics, second by second 10. https://t.co/O5vV9m0R0T — Current number of tweets and searches being posted 11. https://t.co/frkFY1rSbl — Compare the true sizes of countries 12. https://t.co/XuPeoS7yVH — Maps from centuries ago 13. https://t.co/bWn9t1Cptl — Archive of 150,000 historical maps 14. https://t.co/qRf2MHCu0c — World map drawn by volunteers 15. https://t.co/uDVcmXkxMB — Look out the window of a random person around the world 16. https://t.co/h8IoRSYkdh — Virtual walks through cities 17. https://t.co/Vjoyi37Bv9 — Teleport to a random spot on Earth 18. https://t.co/Mg448jMKwt — Catalog of the world's strangest places 19. https://t.co/8rv8a7AiOF — Interactive knowledge experiences 20. https://t.co/1PvqurTLqX — Scale journey from atom to universe (Updated HTML5 link) 21. https://t.co/fRVKGC6Gvc — Explore the solar system in 3D 22. https://t.co/J14So5wWQY — Real sky map in your browser 23. https://t.co/LGRrD9zmYR — NASA's astronomy picture of the day 24. https://t.co/iKHo7UqqTE — NASA's entire visual archive, free 25. https://t.co/aM2rifi60v — Visual articles told through data 26. https://t.co/aZsDiXS3qD — The state of the world with real data 27. https://t.co/X8UrhmWKae — What we mistakenly think we know about the world 28. https://t.co/Cnt7qBp2W6 — Visualizing complex data 29. https://t.co/jmHVof7QKz — World Bank's open data 30. hview source ↗
-
• twitter 1w positiveI used to type through every call. Half listening, half transcribing, and fully missing the point. Then I stopped taking notes completely. My notes got better: 👇 Littlebird has a feature called Meeting Notes. It listens to your calls, transcribes them, and hands you a clean summary with the decisions and the action… I used to type through every call. Half listening, half transcribing, and fully missing the point. Then I stopped taking notes completely. My notes got better: 👇 Littlebird has a feature called Meeting Notes. It listens to your calls, transcribes them, and hands you a clean summary with the decisions and the action items already pulled out. It works on your computer audio, so it does not matter which app the call is in. Zoom, Meet, a browser tab, a phone call on speaker. If you can hear it, it can take the notes. What actually changes: 1/ You look at the person again Eye contact instead of a keyboard. Clients notice this more than you think. 2/ The boring part is already done Decisions, owners, next steps. Written before you close the tab. 3/ The notes are editable You can go in afterwards and add your own thoughts, so it is your document and not just a transcript dump. 4/ It supports 10+ languages Handy if half your calls are not in English. 5/ The notes do not die in a folder This is the real difference. The call becomes part of the memory Littlebird uses later. So weeks after, you can just ask what was agreed and get an answer instead of scrolling a transcript. Most meeting tools give you a summary you read once and never open again. @LittlebirdAI gives you a summary that keeps working. Here is the link if you want your next call to take its own notes: https://t.co/vQvyxy9r99 New users get $20 off the first purchase. The free plan is enough to test it on a couple of meetings this week. #adview source ↗
-
• twitter 1w positiveJev is turning out to be useful for way more than just AI agents. Database filters work great when you know the exact conditions. But what if you want to ask: → Is this article about software engineering? → What topic fits it best? → How technical is it? Usually, you’d have to move the data into application code, c… Jev is turning out to be useful for way more than just AI agents. Database filters work great when you know the exact conditions. But what if you want to ask: → Is this article about software engineering? → What topic fits it best? → How technical is it? Usually, you’d have to move the data into application code, call a model, then write the results back. pg-jev brings that directly into SQL. It’s an open-source Postgres extension with four simple functions: → jev() filters rows in WHERE → jev_prob() returns a probability → jev_choice() picks a label → jev_score() ranks rows by ordered levels No vector columns. No embedding indexes. pg-jev batches your rows, sends them to Jev, and returns typed answers you can filter, sort, and group like normal database results. Basically, you can ask semantic questions directly inside your SQL queries. This is a really interesting direction for combining databases with LLM-powered classification. Worth checking out if you work with Postgres and AI.view source ↗
-
• twitter 1w positiveI see a lot of people liking Grok or Muse and I get it... but tbh the only thing that Grok has that hermes doesn't have yet its a pretty app that's fine but you're locked on a model, muse data mines you and if you want anything new added to it.. you're stuck waiting for their new release that's why I love opensource … I see a lot of people liking Grok or Muse and I get it... but tbh the only thing that Grok has that hermes doesn't have yet its a pretty app that's fine but you're locked on a model, muse data mines you and if you want anything new added to it.. you're stuck waiting for their new release that's why I love opensource and hermes.. it's truly a game changerview source ↗
-
• twitter 1w positiveBut bro, it's like a free $200 gift card! https://t.co/FnekA2gTaB But bro, it's like a free $200 gift card! https://t.co/FnekA2gTaBview source ↗
-
• twitter 1w positiveThis friday is my last day of work. I'm going full time indie dev. I've got ~£100k in savings. • This is many years of runway in SEA. • Doing ~$3-4k/m from my income streams. The first stop is Vietnam 🇻🇳 Then, I'm thinking Cyprus long term 🇨🇾 I decided to go all in abt a year ago. Sharing the journey on x for ~5 mo… This friday is my last day of work. I'm going full time indie dev. I've got ~£100k in savings. • This is many years of runway in SEA. • Doing ~$3-4k/m from my income streams. The first stop is Vietnam 🇻🇳 Then, I'm thinking Cyprus long term 🇨🇾 I decided to go all in abt a year ago. Sharing the journey on x for ~5 months now. It's been an incredible journey so far. But not been easy, I've drifted apart from friends, been stressed and anxious, pushed myself to my limits. But it's the right decision. I've met so many amazing people here, and created a new life for myself, financial freedom. It's all been worth it. We're just getting started 🫡view source ↗