Generative AI

Did ChatGPT ruin em dashes forever? 👀 The em dash used to be a writer’s flex. Stylish. Intentional. A pause with purpose. 🚫Now? It’s a red flag. Readers across Reddit and X are calling it the “ChatGPT hyphen.” It’s become a running joke, and a quiet filter. See a dash, feel a doubt. The wild part? It’s not about grammar. It’s about authenticity. People are second-guessing emotional texts from friends. Editing themselves to avoid looking “AI-written.” Even human writers are swapping out em dashes for commas... just to pass the vibe check. So what now? Cut the em dash? Use it defiantly? Prompt ChatGPT to replace it with commas and periods? 👉Whatever your stance, one thing’s clear: punctuation is no longer neutral. It’s a vibe check. This isn’t just about writing. It’s about how tech is subtly reshaping how we interpret sincerity, trust, and intent. Is the em dash still a power move, or has it been auto-corrected out of human expression? 👀 Built an AI tool? Get it featured in our community of 12M+ https://hubs.li/Q03B2_t60 #chatgpt #openai #artificialintelligence #llm #machinelearning

How long did it take before you realized this was AI? The casual walk toward the CCTV and the pose like old friends must have been a give away right?😂 It feels real. But it's not. It's AI. What once needed a Hollywood budget is now doable with a prompt and the right model. Realism isn’t just closing the gap with reality, it’s blurring it entirely. This isn’t about entertainment alone. It’s a wake-up call for: ✅Security: Deepfake CCTV can seed misinformation. ✅Trust: Visuals no longer equal truth. ✅Creative power: Anyone can now produce surreal storytelling at scale. As AI-generated video gets better, faster, and cheaper… The real question is: What systems do we need to build before we no longer trust our own eyes? Watch GenAI come alive on TikTok with real use cases and insights https://hubs.li/Q03B2TSt0 Video: riooooo. xyz (via IG) #generativeai #veo3 #aiart #googledeepmind #gemini

Now, this is a robotics application we can all get behind.🤖 The Figure 01 humanoid just did laundry: fully autonomously. Not tele-operated. No scripts. No task-specific fine-tuning. It runs on Helix, Figure AI’s Vision-Language-Action model, combining: 👉A 7B vision-language model for high-level planning 👉An 80M visuomotor controller for real-time precision The result? F.02 can fold towels, sort laundry, and generalize to novel objects in real homes, something that’s been out of reach for robotics until now. - And it all runs onboard, in real time, powered by low-watt GPUs. - No cloud latency. No remote compute. This is what moving from robotics demos to real-world deployment looks like. So be honest: would you trust a robot like this alone in your house? Deploy your own MCP server and unlock real-world GenAI workflows https://hubs.li/Q03B2S2f0 #humanoid #robotics #artificialintelligence #machinelearning #algorithm

🎬 AI just gave Titanic a rewrite, and Jack lives. The internet’s loving it, but should we? Sure, it’s fun. But let’s not ignore the cost. Sad endings, tough choices, and unresolved tension are what made stories like Titanic timeless. The heartbreak was the point. When AI starts airbrushing away the grief, the loss, the messy human moments… are we flattening the very thing that makes art powerful? This isn’t just about a meme. It’s about a future where machines might default to “feel-good” over “feel real.” But here’s the opportunity: The more AI reshapes our shared stories, the more we need clarity, guidance, and ethical frameworks that scale with the tech. Video: Anson Kao 👉 In an AI-assisted world, who protects the rawness that makes art human? Subscribe on YouTube to get weekly GenAI walkthroughs, real‑world agent demos, and expert insights https://hubs.li/Q03B2QXl0 #generativeai #artificialintelligence #aiart #aiethics #titanic

OpenAI’s next model is already solving problems that left senior engineers stuck. Meanwhile, Meta is retooling its AI ambitions into wearable personal superintelligence, NASA is testing autonomous AI robots aboard the ISS, and Neuralink has helped a paralyzed woman write her name again after 20 years. Neural, wearable, autonomous. Each new interface advances the disassembly of human exclusivity. Read on. Your perspective belongs in the thread. 👇 #ai #tech #openai #nasa #meta

The 21 AI tools powering productivity, sales, and beyond New month, fresh stack. These AI stacks are helping teams across functions unlock better workflows, sharper decisions, and more output with less drag. Security • Tumeryk https://lnkd.in/gg6i49ps • Darktrace • Opaque AI Infrastructure • Plaud AI https://lnkd.in/gRDZCxXC • Otter. ai • Fireflies. ai Productivity • Quillbot • Taskade • Microsoft 365 Copilot AI Agent • ChatGPT • Perplexity • Grok Video • Guidde • Midjourney • Sora Sales • HubSpot • Salesforce • Klaviyo Image • Adobe Firefly • DALL·E • Leonardo — Want your tool featured in the next edition? 📣 Submit to our 12M+ community → https://lnkd.in/gctriXTq

𝐖𝐡𝐲 𝐛𝐮𝐢𝐥𝐝𝐢𝐧𝐠 𝐥𝐚𝐧𝐠𝐮𝐚𝐠𝐞 𝐦𝐨𝐝𝐞𝐥𝐬 𝐢𝐧 𝐮𝐧𝐝𝐞𝐫𝐫𝐞𝐩𝐫𝐞𝐬𝐞𝐧𝐭𝐞𝐝 𝐥𝐚𝐧𝐠𝐮𝐚𝐠𝐞𝐬 𝐦𝐚𝐭𝐭𝐞𝐫𝐬? Join us for a special live session with Christos Petrocheilos, a recent graduate of our AI Executive Program, as he presents his pioneering work developing a Large Language Model (LLM) in Greek. In this conversation, Christos will walk us through the technical and linguistic journey of creating an LLM tailored to a specific language and culture, and why that matters for the future of inclusive AI. This isn’t just a project demo. It’s a case study in how local innovation can shape global conversations in AI. 𝐖𝐡𝐨 𝐢𝐭’𝐬 𝐟𝐨𝐫: AI researchers, engineers, linguists, and anyone passionate about responsible, multilingual AI development. 𝐖𝐡𝐚𝐭 𝐲𝐨𝐮’𝐥𝐥 𝐠𝐞𝐭: Insights from Christos’ real-world build, lessons learned from the process, and what it means to localize large-scale AI. 𝐁𝐨𝐧𝐮𝐬: This session is part of our AI Executive Education Series - showcasing exceptional alumni and the projects they’re bringing to life.

⚠️ Heads-up: Your “private” ChatGPT chats might be showing up on Google. You’ve probably shared a chat or two using OpenAI’s share feature, maybe with your team, maybe on Slack, maybe even in a workshop or with friends! Those links are public. And they’re being indexed by search engines. 👀 This means: – Your internal prompts? Searchable. – That early-stage product idea? Crawled. – That clever GPT workflow you shared with friends? Potentially exposed. It’s called Google dorking: a known way to find data not meant to be public. And AI chats are now part of the mix. 🔒Privacy tips for AI users: 1. Think twice before hitting Share. It’s not as private as it sounds. 2. Don’t include sensitive info: names, emails, company docs, etc. 3. Disable memory or training when working on confidential tasks. 4. Set internal AI usage policies that include this. We’re entering a new era where visibility is default. Oversharing could be your biggest risk, not your model. Here’s the big Q: Should OpenAI (and others) be responsible for preventing AI chat indexing, or is this on us to control? 👇 Let’s talk. Choose your interests and get curated GenAI updates that matter https://hubs.li/Q03zW-Hq0 #artificialintelligence #chatgpt #privacy #openai #aiethics

👀OpenAI officially lands in Europe with Stargate Norway, and brings 100K GPUs with it. A massive AI data center built in Narvik is not just about GPUs, it’s a geopolitical move. Here’s everything you need to know👇 ➡️Location is a power play: Cool climate. Cheap hydropower. Full-on renewable energy. ➡️100,000 NVIDIA GPUs by 2026: That’s not a data center. That’s a compute fortress. ➡️Homegrown devs get first dibs: Finally, infra that serves local talent, not just Silicon Valley. ➡️Waste heat powers nearby businesses: Sustainable and circular. OpenAI's Stargate in Narvik may be a big win for Norway, but let’s not forget the bigger picture. With 100K GPUs, it’s more than just tech, it’s about power and influence. But will this be a sustainable model, or just another data center chasing the next trend? 🤔 Will your country build its own Stargate, or rent from someone else’s? Built an AI tool? Get it featured in our community of 12M+ https://hubs.li/Q03zNQfB0 #artificialintelligence #openai #nvidia #sovereignty #openai #chatgpt #stargate

Top Companies Defining the Future of AI | Q2 2025 In this special edition of the Atlas AI Newsletter (read by 3M+ builders, operators & decision-makers), we spoke with C-level execs behind today’s most influential AI platforms—spanning LLMs, RAG frameworks, AI agents, and enterprise infrastructure. We asked them the hard questions: 💡 What core problem do you solve? 🔍 What sets your tech apart? 🏆 What’s your biggest 2025 win? 🔮 How are you shaping the future of AI? Here are 5 companies redefining what’s possible: Wand AI — The OS for AI-human collaboration. AI agents and humans working side-by-side—with full AgentOps, enterprise control, and deployment flexibility. Vectara— Enterprise-grade RAG without hallucinations. Factual consistency, hallucination correction, and full-stack evaluation tools for building trusted AI agents. Runbear — Your AI teammate for ops-heavy teams. Used by BCG, Peek & LaserAway. Runbear auto-scans work, identifies inefficiencies, and completes tasks via 2,700+ integrations. Quickchat AI — Conversational commerce, powered by AI. Chat with a Shopify store like it’s a sales assistant. Browse, shop, and checkout—all within the conversation. LiveX AI — Memory-aware agents for CX, retention & growth. Powering support, onboarding, and sales for brands like Wyze, FashionPass & Pictory. — Built an AI tool? Get it featured in our community of 12M+ https://lnkd.in/dagGawdv #AI #GenAI #AIagents #LLMs #RAG