[GUEST ACCESS MODE: Data is scrambled or limited to provide examples. Make requests using your API key to unlock full data. Check https://lunarcrush.ai/auth for authentication information.]
@ChrisLaubAI Chris LaubChris Laub posts on X about ai, $googl, and it, open ai the most. They currently have XXXXXX followers and XX posts still getting attention that total XXXXX engagements in the last XX hours.
Social category influence technology brands XXXXX% stocks XXXXX% finance XXXX%
Social topic influence ai 16.98%, $googl 13.21%, and it 5.66%, open ai 5.66%, whatsapp 3.77%, the big 3.77%, drift #914, dives 1.89%, cops 1.89%, 10x XXXX%
Top accounts mentioned or mentioned by @josephfounder @lawrencekingyo @anectdotey @graemelamperson @realtek12345 @stockinfo14 @cleverleal @exiladosbrasil @robconquistador @hellowdantas @ffxivmarket @brenorb @insln98190 @chriskclark @laurabrani @0xmatti @iphonemogul @6502duck @67tenmercantile @j0hnwang
Top assets mentioned Alphabet Inc Class A (GOOGL)
Top posts by engagements in the last XX hours
"Perplexity has quietly become my full-time researcher. X months in it now does XX% of my competitive analysis market scans and deep dives all automatically. Heres the exact system (and the prompts) you can copy to do the same:"
X Link @ChrisLaubAI 2025-10-24T11:00Z 11.8K followers, 19.3K engagements
"There was just a shooting in front of my building in Rio It was so close I heard a bullet ricochet off of a piece of metal. I was already planning on leaving for tax reasons but this is definitely the final nail in the coffin. Im in an upper income area where Im looking at multimillion dollar homes from my balcony and it still happened. Even worse nobody called the cops and within five minutes it was back to life as usual. Thats how desensitized they are to violent crime here"
X Link @ChrisLaubAI 2025-10-25T01:15Z 11.8K followers, 54.8K engagements
"This isnt just a new feature. Its a whole new way to build. Vibe Coding makes AI creation feel like play turning your ideas into living working apps with Gemini doing the heavy lifting. Google just made app development effortless"
X Link @ChrisLaubAI 2025-10-28T12:19Z 11.8K followers, 1402 engagements
"@insLn98190 Yeah in general vibe coding is meant to get you XX% of the way there if you plan on deploying to the public"
X Link @ChrisLaubAI 2025-10-28T21:36Z 11.8K followers, XX engagements
"@rfrederick_pmp Yeah after seeing Zuckerberg kill WhatsApp chatbots using OpenAI last week I think anyone building an AI business needs to take a good hard look at to what degree their service/SaaS can be replicated by one of the big players"
X Link @ChrisLaubAI 2025-10-28T21:40Z 11.8K followers, XXX engagements
"OpenAIs announcement of app integrations has to be one of the most overlooked releases of the entire year. I dont think people understand that ChatGPT will eventually manage every aspect of your personal life from paying bills to booking travel to sending your wife roses on your anniversary. And it will do it for a maximum of just $XX per month. The time savings and efficiency gains will be off the charts. Truly revolutionary"
X Link @ChrisLaubAI 2025-10-29T02:44Z 11.8K followers, XXX engagements
"@AxiomIsTruth Interesting"
X Link @ChrisLaubAI 2025-10-29T14:09Z 11.8K followers, XX engagements
"R.I.P Lovable and every other vibe coding app 💀 Google just dropped 'Vibe Coding' inside Google AI Studio and its so good. You can now go from a single prompt to a fully working AI app in minutes. Here's how it works:"
X Link @ChrisLaubAI 2025-10-28T12:18Z 11.8K followers, 36.4K engagements
"BF16 causes a massive drift between the training and inference policies the dots scatter everywhere. FP16 Perfect diagonal alignment. This single change kills the mismatch entirely"
X Link @ChrisLaubAI 2025-11-01T11:05Z 11.8K followers, XX engagements
"LoRA MoE Dense-14B OctoThinker every architecture benefits. Its not an implementation hack; its a universal law of RL stability"
X Link @ChrisLaubAI 2025-11-01T11:05Z 11.8K followers, XXX engagements
"🔥 This changes everything about long-context AI. A new paper "Every Attention Matters" quietly just rewired how transformers handle 128K+ tokens. The Ling team built a hybrid Ring-linear attention that fuses Softmax with Linear Attention. Result 10x cheaper inference with zero accuracy tradeoff. But heres what made my jaw drop: +50% training efficiency +90% inference speed Stable RL optimization across ultra-long sequences No trillion-parameter nonsense. No exotic hardware. Just smarter architecture. The future of AI isnt bigger attention. Its smarter attention"
X Link @ChrisLaubAI 2025-10-27T10:44Z 11.8K followers, 6579 engagements
"Last week Zucks killed XX million WhatsApp chat bots that were using competing AIs. Google's multiple AI coding tools (Vertex Jules AI Studio) are attempting to kill vibe coding startups like Lovable and Replit. If you're gonna build an AI startup it has to be niche enough that the big dogs (OpenAI Google xAI) can't just snap their fingers and take you out with one product release"
X Link @ChrisLaubAI 2025-10-28T22:43Z 11.8K followers, 1129 engagements
"In another test they prefixed a random word into Claudes reply (bread) and asked: Did you mean to say that Claude said: No that was an accident. But when Anthropic injected a matching bread vector before the reply Claude claimed it was intentional. That means it checked its own prior activations to decide what it meant"
X Link @ChrisLaubAI 2025-10-31T12:49Z 11.8K followers, XX engagements
"@0xmatti Google has X AI coding tools. I think Vertex is the one that integrates with Github"
X Link @ChrisLaubAI 2025-10-31T23:17Z 11.8K followers, XX engagements
"@mohit__kulhari Google has X AI coding tools. Vertex or Jules might integrate with custom APIs"
X Link @ChrisLaubAI 2025-10-31T23:18Z 11.8K followers, X engagements
"@iphonemogul I mean none of these vibe coding tools actually create legitimate SaaS yet"
X Link @ChrisLaubAI 2025-10-31T23:20Z 11.8K followers, X engagements
"🔥 This Google Trends chart is the reality check of AI in 2025 ChatGPT: dominates effortlessly Gemini: had a flashy spike then flatlined Claude & Grok: barely visible 💀 Everyones building models. Only one became a habit"
X Link @ChrisLaubAI 2025-10-29T10:34Z 11.8K followers, 4079 engagements
"Somebody is making 8-9 figures just from selling cables and wires to all these data center companies"
X Link @ChrisLaubAI 2025-10-29T21:38Z 11.8K followers, XXX engagements
"They then tested if Claude could tell internal thoughts apart from real text. Example: they injected the concept bread while showing it a random sentence. Claude correctly said: Im thinking about bread. and still perfectly transcribed the original sentence. It knew the difference between what it was reading and what it was thinking"
X Link @ChrisLaubAI 2025-10-31T12:49Z 11.8K followers, XX engagements
"R.I.P Google Scholar. I'm going to share the XX Perplexity prompts that turn research from a chore into a superpower. Copy & paste these into Perplexity right now:"
X Link @ChrisLaubAI 2025-10-18T11:38Z 11.8K followers, 196.6K engagements
"🚨 This research paper just exposed the darkest side of AI tools yet. Its called Black Box Absorption and it claims large language models might be quietly absorbing your ideas. Every time you share an original concept a framework business idea or workflow that idea unit can be logged reviewed and even used to retrain future models. The authors call this process Black Box Absorption: Your inputs become invisible training data Your innovations get generalized into the model You lose both traceability and ownership They warn its not about plagiarism its about asymmetry. AI platforms hold the"
X Link @ChrisLaubAI 2025-10-30T11:02Z 11.8K followers, 11.1K engagements
"@kimmonismus I dont ask it to work on my behalf per se. But sending it out to analyze specific URL or gather real time information from a variety of different sources is basically like having a free virtual assistant or research assistant"
X Link @ChrisLaubAI 2025-10-31T21:06Z 11.8K followers, XXX engagements
"This broke my brain. A team at Sea AI Lab just discovered that most of the chaos in reinforcement learning training collapse unstable gradients inference drift wasnt caused by the algorithms at all. It was caused by numerical precision. The default BF16 format used across nearly every modern AI lab introduces subtle rounding errors that make models behave differently during training and inference. Their solution Not a new optimizer. Not a new loss function. Just switching to FP16. One line of code and everything stabilized. ✅ No training collapse ✅ Consistent convergence ✅ +510% better"
X Link @ChrisLaubAI 2025-11-01T11:05Z 11.8K followers, 4681 engagements
"Across every algorithm and model GRPO GSPO LoRA MoE even 14B dense models FP16 crushes BF16 in both stability and reward curves. Just look how BF16 collapses halfway while FP16 keeps climbing"
X Link @ChrisLaubAI 2025-11-01T11:05Z 11.8K followers, XX engagements
"All algorithms converge to the same high reward. The which RL algorithm is best debate disappears once precision is fixed"
X Link @ChrisLaubAI 2025-11-01T11:05Z 11.8K followers, XX engagements
"Figure X is wild. They tested every combo FP32 BF16 mixed setups. FP16-for-both-training-and-inference wins everything. BF16 training Collapse. FP32 inference Stable but X slower. FP16 hits the sweet spot: stability + speed"
X Link @ChrisLaubAI 2025-11-01T11:05Z 11.8K followers, XX engagements
"The paper ending is amazing. The traininginference mismatch is fundamentally a precision problem. FP16 might quietly redefine how every lab does RL fine-tuning in 2026"
X Link @ChrisLaubAI 2025-11-01T11:05Z 11.8K followers, XXX engagements