Nikhil Gupta’s Post

I keep reading that “AI progress is slowing down”. But it was 6 years from the paper “Attention is all you need” until ChatGPT was released. That’s 6 years of incremental updates to get to an AI model that regular consumers would find useful. ChatGPT is now 2 years old. Maybe it only seems like incremental improvements in that time. But I’ve seen plenty of use cases where the models finally get good enough to be useful. Not only that but we’re getting better at using them. “But aren’t we running out of training data?” No. No we are not. The big players in AI have all sorts of ways of getting more data. That’s not the biggest problem right now. The biggest problem is that additional data is only giving incremental gains. But everyone knows that. It’s not some secret. New papers come out every week about how to get more out of these models. It also seems to take more researchers, more engineers, and more compute to make each incremental update. That’s because everyone is focused on one paradigm and we keep finding ways to squeeze more juice out of it. Paradigm shifts happen on slower time scales while incrementalism gets less and less effective. But incrementalism is how we discover paradigm shifts. “Attention is all you need” was an insight that came from trying to squeeze more performance out of the last paradigm.

 🦾 𝗖𝗹𝗶𝗰𝗸𝗨𝗽 𝗔𝗜 𝗝𝘂𝘀𝘁 𝗚𝗼𝘁 𝗦𝗺𝗮𝗿𝘁𝗲𝗿 – 𝗦𝗮𝘆 𝗚𝗼𝗼𝗱𝗯𝘆𝗲 𝘁𝗼 𝗦𝘄𝗶𝘁𝗰𝗵𝗶𝗻𝗴 𝗕𝗲𝘁𝘄𝗲𝗲𝗻 𝗧𝗼𝗼𝗹𝘀! Are you still using ChatGPT or other similar AI tools? You don't need to! With our latest updates, ClickUp AI has everything you need: 1️⃣ 𝗔𝘀𝗸 𝗔𝗜 𝗮𝗯𝗼𝘂𝘁 𝗨𝗽𝗹𝗼𝗮𝗱𝗲𝗱 𝗙𝗶𝗹𝗲𝘀 Ask AI to explain a picture or analyze a document. Just upload it! 2️⃣ 𝗨𝗽𝗱𝗮𝘁𝗲𝗱 𝗪𝗿𝗶𝘁𝗶𝗻𝗴 𝗧𝗼𝗼𝗹𝘀 Get an improved draft or specific edits, with a single click. 3️⃣ 𝗖𝗼𝗻𝘃𝗲𝗿𝘀𝗮𝘁𝗶𝗼𝗻 𝗛𝗶𝘀𝘁𝗼𝗿𝘆 Pick up a previous discussion with Brain, right where you left off! We’ve also taken your feedback to heart: Beyond these new features, we’ve improved the overall quality and reliability of ClickUp AI throughout the platform. Check it out! 📈 Your workday, simplified. Your results, amplified. Try the new AI tools today! 👉 https://bit.ly/49u8sQn

I recently asked my friend ChatGPT what trends it was seeing for using AI in recognition. The top 5 trends listed were: 1. Personalized Recognition 2. Real-time Feedback 3. Data Analytics for Insights 4. Integration with HR Systems 5. Predictive Analytics While we are currently exploring ways to integrate AI in our flow of work with more robust system integrations, predictive analytics and data insights, we haven’t yet made the proverbial leap into leveraging it holistically in our recognition platforms for writing and creating more personalized recognition experiences. Curious, how are other organizations leveraging it today? What are your thoughts on the use of AI in recognition?

If your AI stack is Chat-GPT + RAG, this one's for you! 👋🏻 You don't own your model. As you grow, your IP is your model and you don't own it. It's like ordering takeout and claiming it as your own food to serve to guests - it doesn't scale. 🥡 🤖 Hallucinations will be a problem. There is no substitute for knowledge injection in your model, and that won't happen with this setup. 💸 Your costs will eventually exceed your value. RAG is expensive. Is RAG needed in all use cases? No. It's there to ensure your most recent data is remembered, but what about historical data? Catastrophic forgetting is real. 📊 Own your model, use your data, and drive the right mix of knowledge injection and remembrance with continual pretraining for your use case. 🔧 There are alternatives. Arcee.ai 🤙🏻 #AI #MachineLearning #TechInnovation #DataManagement #AIModels

What if your searches could think like you do? ChatGPT’s advanced search capabilities, like SearchGPT, allow you to ask detailed, complex questions and receive targeted, actionable answers—without wasting time sifting through irrelevant results. At Your AI Wizards, we’ve begun using SearchGPT to dig deeper into trends, uncover the latest insights, and find the answers we need—fast. And we’re impressed! For small businesses, staying ahead of evolving technology is essential. That’s why we’re excited to introduce our new tool, AI In Site. It’s designed to help you navigate the changing world of search with confidence and precision. The future of online search is already here—are you ready? Visit http://www.aiinsite.io to learn more, and let’s explore how next-generation search can work for your business. #SearchGPT #AIInSite #AIForBusiness #YourAIWizards

🚀 Excited to introduce my latest innovation in AI technology: 𝐄𝐥𝐚𝐫𝐚𝐁𝐫𝐢𝐞𝐟𝐢𝐟𝐲! 🚀 𝐄𝐥𝐚𝐫𝐚𝐁𝐫𝐢𝐞𝐟𝐢𝐟𝐲is an innovative AI-based web application developed by me (I have used Chatgpt-4 model here), designed to revolutionize the way we consume information. With 𝐄𝐥𝐚𝐫𝐚𝐁𝐫𝐢𝐞𝐟𝐢𝐟𝐲, you can effortlessly condense lengthy news articles and website content into concise summaries, providing quick access to key insights and saving valuable time. Whether you're staying updated on current events or researching various topics, 𝐄𝐥𝐚𝐫𝐚𝐁𝐫𝐢𝐞𝐟𝐢𝐟𝐲streamlines the process, offering clarity and efficiency in information digestion. Experience the power of summarization with 𝐄𝐥𝐚𝐫𝐚𝐁𝐫𝐢𝐞𝐟𝐢𝐟𝐲 today. 𝐋𝐢𝐯𝐞 𝐋𝐢𝐧𝐤 : 𝐡𝐭𝐭𝐩𝐬://𝐞𝐥𝐚𝐫𝐚𝐛𝐫𝐢𝐞𝐟𝐢𝐟𝐲.𝐧𝐞𝐭𝐥𝐢𝐟𝐲.𝐚𝐩𝐩/ 𝐻𝑜𝑤 𝑡𝑜 𝑢𝑠𝑒 𝐸𝑙𝑎𝑟𝑎𝐵𝑟𝑖𝑒𝑓𝑖𝑓𝑦: 1️⃣ 𝐶𝑜𝑝𝑦 𝑡ℎ𝑒 𝑙𝑖𝑛𝑘 𝑜𝑓 𝑎𝑛𝑦 𝑎𝑟𝑡𝑖𝑐𝑙𝑒-𝑏𝑎𝑠𝑒𝑑 𝑤𝑒𝑏𝑠𝑖𝑡𝑒, 𝑠𝑢𝑐ℎ 𝑎𝑠 𝑎 𝑛𝑒𝑤𝑠𝑝𝑎𝑝𝑒𝑟. 2️⃣ 𝑃𝑎𝑠𝑡𝑒 𝑡ℎ𝑒 𝑙𝑖𝑛𝑘 𝑖𝑛𝑡𝑜 𝐸𝑙𝑎𝑟𝑎𝐵𝑟𝑖𝑒𝑓𝑖𝑓𝑦. 3️⃣ 𝐼𝑛𝑠𝑡𝑎𝑛𝑡𝑙𝑦 𝑟𝑒𝑐𝑒𝑖𝑣𝑒 𝑎 𝑠𝑢𝑚𝑚𝑎𝑟𝑖𝑧𝑒𝑑 𝑣𝑒𝑟𝑠𝑖𝑜𝑛 𝑜𝑓 𝑡ℎ𝑒 𝑒𝑛𝑡𝑖𝑟𝑒 𝑎𝑟𝑡𝑖𝑐𝑙𝑒. Experience the power of AI-driven summarization with 𝐄𝐥𝐚𝐫𝐚𝐁𝐫𝐢𝐞𝐟𝐢𝐟𝐲 and transform the way you consume information. #AI #Innovation #ElaraBriefify #ContentSummarization #Efficiency #Productivity

𝗖𝘂𝘀𝘁𝗼𝗺𝗚𝗣𝗧𝘀: 𝗧𝗵𝗲 𝘀𝗲𝗰𝗿𝗲𝘁 𝗔𝗜 𝘄𝗲𝗮𝗽𝗼𝗻 𝗳𝗼𝗿 𝗯𝗼𝗼𝘀𝘁𝗶𝗻𝗴 𝗮𝗹𝗹 𝗼𝗳 𝘆𝗼𝘂𝗿 𝗺𝗼𝗻𝗲𝘆-𝗺𝗮𝗸𝗶𝗻𝗴 𝗼𝗽𝗽𝗼𝗿𝘁𝘂𝗻𝗶𝘁𝗶𝗲𝘀 Wendy Breakstone calls BS on customGPTs being old news. She also feels like "Templates are so 2021." 😉 Yep, you heard that right! I love how Wendy’s not afraid to shake things up. Wendy Breakstone, AI Educator & Marketing Strategist, shared with us at the AI Made Easy Summit exactly what customGPTs are and how she uses them to help her clients level up in their business in record speeds. "CustomGPTs don’t just enhance your programs—they create a ripple effect that significantly boosts your bottom line. These will be your secret weapon for boosting all of your money-making opportunities!" Wendy shares. So, what exactly are custom GPTs? They’re personalized versions of ChatGPT, fine-tuned with specific instructions to perform set tasks for your business. They save you loads of time by automating repetitive work and consistently delivering the results you need. And the best part? Anyone can build them . Wendy made it clear—it’s not about being tech-savvy. It’s about knowing how to communicate with the GPT effectively to create a tool that works for you. Wendy's top tip is to get ideas first. Look for the places your clients tend to get stuck in your programs—that’s where custom GPTs can help them reach success faster. Curious to learn more? 1. Watch this clip from Wendy's talk on customGPTs at the AI Made Easy Summit. 2. While you’re there, don’t forget to grab Wendy's free guide: "75+ Ways to Use CustomGPTs To Make You Money." 4. To watch the entire training plus 10 more, get Unlimited Access to the AI Made Easy Summit here: https://lnkd.in/gFzfvuwX 🔥 #AI #AIMADEEASY

🤓 As an AI geek, I get excited when a trusted source like WSJ shares actionable insights minus the hype. 🤖 Why does it excite me? Because, for example, it’s in a language even my dad can understand. I can talk to him about Chat GPT until his eyes glaze over. But if he reads it in print for WSJ? “Why of course son, everyone knows how huge this is!” Things change quickly, but this article is an excellent snapshot in time. Here are our top insights we fully agree with and will be quoting as reference moving forward: 1️⃣ Treat your AIs like employees Context. Context. Context. In the same way you'd orient a new hire, guide ChatGPT-4 through your company's communication style: prioritizing clarity over aggressive marketing.   2️⃣ No one has ever gotten fired for buying Chat GPT Darned if that’s not a pithy summary statement from the writers at WSJ. We’ve been touting Chat GPT with its access to APIs for non-enterprise businesses. It’s not just AI, it potentially a full business operating system. We try to avoid cliche. But after hearing hundreds of times from a wide variety of sources, we can’t wrap it up any better: “It’s the best $20 you’ll ever spend.” 3️⃣ Figure our what matters for your AI  AI is not an off-the-shelf cure-all solution. You must identify where it can make the most significant impact on your operations. Often, that means deploying it on the time-consuming, mundane tasks—think processes and admin work that drain your hours. The true power of AI in business lies in its thoughtful integration into your daily operations. Steer clear of tech for tech's sake. It's about making AI work for you in the most practical sense. So there you have it - wisdom from the WSJ distilled. We’ve already integrated these insights into our roadmap and can't wait to see where they’ll lead us and our clients. What about you? Have you found a unique way to employ AI in your business that defies the norm? Drop your thoughts and experiences below 👇👇👇 #BusinessAI #FutureOfWork #chatgpt4 Link to source article, but it's behind a paywall: https://lnkd.in/gR69nB67

I hear alot about GenAI automation for sales and marketing...  "Look what I did with GPT. It's fast, so smart, and saves time". But remember, AUTHENTICITY is required to build trust. I see those of you using automation... my response: "Hey, your AI is showing!" BTW - Here's what ChatGPT thinks I should have said: I keep hearing about GenAI automation for sales and marketing… “Check out what I achieved with GPT—it’s fast, intelligent, and a huge time-saver.” But let's not forget, authenticity is essential for building trust. I can spot those relying on automation, and my reaction? “Your AI is showing!” See the difference? Great article that echos our sentiments.  https://lnkd.in/emSG2895 #SalesStrategy.

Why CoT Works: Iterative Computation for Serial Problems Ever wondered when to use fancy Chain of Thought (CoT) prompting, and when regular ChatGPT will do? Here's the key: Serial problems NEED CoT. Turns out, without CoT, we can only solve problems that have fast, parallel solutions. Let's get practical: - Travel Planning: Imagine an AI travel agent. CoT lets the AI figure out your destination, considering your preferences and budget. Then, it factors in dates (weather and your schedule), books your stay, and suggests activities based on the destination's attractions and your interests – each step builds on the last, ensuring a perfect trip. This is a classic example of a serial problem – each step depends on the outcome of the previous one, making it impossible to ask five people to do these steps in parallel. - Sentiment Analysis: This one ISN'T serial. This is a parallel task! A regular transformer can analyze a review for positive/negative/neutral vibes all at once. Imagine it like asking multiple people to independently assess the sentiment of the text simultaneously. They don't need to know the results of each other's analysis or follow any specific order. The Tech Deep Dive (Simplified!) CoT basically lets transformers tackle way more complex problems. Without CoT, they're limited to AC0 problems (solvable by constant-depth circuits). But add CoT, and now we're talking problems solvable by log-depth circuits (NC1 complete problems). Want proof? Check out the paper "Chain of Thought Empowers Transformers to Solve Inherently Serial Problems" (https://lnkd.in/giCaSGA5). Author: Zhiyuan Li, Hong Liu, Denny Zhou , and Tengyu Ma The Takeaway: Next time you've got a problem for your transformer, ask: Is this serial (step-by-step), or can it be done in parallel? That'll tell you if you need the power of CoT. #ChainOfThought #Transformers #CoT #LLM #AI