
You need to read the third annual AI Adoption Report from Wharton Human-AI Research. ↓↓↓ - 82% use generative AI weekly (up from 72% in 2024 and 37% in 2023) - 46% use generative AI daily (+17pp YoY) - nearly 75% already see positive ROI from generative AI; 4 out of 5 see generative AI investments paying off in 2-3 years - 88% anticipate generative AI budget increases in the next 12 months; 62% anticipate a more than 10% increase in the next 2-5 years - IT and purchasing/procurement lead on AI use and confidence (Allie note: IT makes sense to me, procurement does not). Marketing, Sales, Ops lag behind on adoption (Allie note: Marketing and Sales does not make sense to me) - 72% report tracking structured, business-linked ROI metrics, like profitability, throughput, and productivity (Allie note: much lower than it should be) - VP and above are more optimistic on AI ROI (81% versus mid-managers at 69%) - Chief AI Officer roles are now present in 60% of enterprises (Allie note: I really don't think this should be a role unless you're in a services org like Accenture, where AI is a business line) - Only 61% (+7pp YoY) are implementing training and awareness programs and investment in training has dropped 8pp YoY. This, coupled with the stat from AWS that upskilling is one of the biggest blockers in the enterprise, makes me less hopeful for enterprise-wide change. Methodology is on page 4 of the report in detail. It was a 15min online survey. Total respondents in 2025 is ~800, 2024 had ~800, 2023 had 670. Roles are senior decision makers in HR, IT, Legal, Mktg, Sales, Product, Eng, Finance, Mgmt. 📌 MORE IN THREAD - save this thread for your next business presentation
Writing about where AI-powered software is going and I’ve come up with a new name for what we’re all going to end up building. Essay coming later today.
On that note, I am currently running a large-scale experiment on the upcoming inference-scaling chapter: A) Parallel Sampling - Self-Consistency (Majority Vote) - Rejection Sampling - Best-of-N (Verifier-Based) B) Sequential Refinement - Self-Refinement - Power Sampling - MCMC (Simple) - MCMC (Block as in "Reasoning with Sampling" paper) - Tree-of-Thought ......
your screen now understands how you move in virtual worlds

🎙️ New article published: “𝗜𝗻𝘀𝗶𝗱𝗲 𝗟𝗮𝗿𝗴𝗲 𝗟𝗮𝗻𝗴𝘂𝗮𝗴𝗲 𝗠𝗼𝗱𝗲𝗹𝘀: 𝗧𝗵𝗲 𝗔𝗜 𝗧𝗵𝗮𝘁 𝗨𝗻𝗱𝗲𝗿𝘀𝘁𝗮𝗻𝗱𝘀, 𝗦𝗽𝗲𝗮𝗸𝘀, 𝗮𝗻𝗱 𝗖𝗿𝗲𝗮𝘁𝗲𝘀” At the core of today’s most powerful AI applications are Large Language Models (LLMs). We use them daily—often without realizing it—but how many truly understand what powers...
Almost every brand will need to rebuild their marketing funnel from the ground up. Alex Sherman, CEO of @Bluefish_AI, explains how brands are adapting to AI: “Marketers have always split budgets between organic and paid, but AI is forcing a rethink. What’s...

Great to see the shoutout from Jensen @nvidia for open-source AI & @huggingface. Open-source is the foundation of AI and we need much more of it in the US & around the world! https://t.co/kFtBAPbqZi
AI is a multiplier, not a miracle. Your data makes or breaks it.
Jony Ive is having to eat his words, I learned from an industry leader last night who has inside connections to @OpenAI. Can't disclose who it is, but he says that OpenAI is readying a full court consumer products push...
If an AI tool creates motion without movement, it’s research. When it compresses a release cycle, trims spend, or expands revenue, it becomes infrastructure.
Robot breakthrough. This is why the 12,000 workers at @Tesla's Fremont Factory are so important. I hear groups of them are training the robots, doing tasks over and over to give its simulator enough data to help it build very advanced...
Best face swap method in ai right now? Progress got nuked during the deepfake drama and I have about 6 minutes of generated footage with a slightly off face that needs a reliable fix
I'm hearing the same. That we are in a lull right now and that breakthroughs are coming. I'm uploading two videos with @blevlabs right now. He has one, and there are others. The videos are long, but worthy to see...

Just saw the MiniMax-M2 benchmarks, and the performance is too good to ignore :). So, I just amended my "The Big LLM Architecture Comparison" with entry number 13! 1️⃣ Full attention modules: As shown in the overview figure below, I grouped...
A huge robotics bubble forming? Learned last night at the @1x_tech party from an investor that @UnitreeRobotics is preparing a public offering, but that USA investors are being rebuffed. The guy I talked to wanted to put in $200 million and...
In this AI Chitchat episode, we sit down with Fernando from Liquid AI to unpack the rise of small language models (SLMs) and why architecture + data now matters more than raw parameter counts. We cover hybrid transformer/convolutional designs, neural...
Researchers at Beijing Academy of Artificial Intelligence (BAAI) trained a Unitree G1 to pull a 1,400 kg car. https://t.co/uC9Xxw5Jap #unitreeg1 #Robotics #humanoid #AI @RLDI_Lamy @Ronald_vanLoon @baski_LA @labordeolivier @TysonLester
Jakub and I are going to do a livestream and answer questions today at 10:30 am pacific. We have a lot of things to talk about--of course we will cover our new corporate structure, but we will also discuss our new...
1/ Hey Brazil, ChatGPT Go is now available for just R$39,99 per month — a more affordable way to use our advanced AI tool. https://t.co/UcfuEreIhg

🚀 The Future of Enterprise Transformation Starts Now 🔍 Harnessing Agentic AI and Model Context Protocol (MCP) for Sustainable Growth 🔹 Businesses are entering a new era driven by autonomy, security, and intelligent adaptability. 🔹 Agentic AI and MCP are redefining...
OpenAI conversion to PBC https://t.co/3tdPRWR4DU
🇺🇸 Amazon has more than 750,000 #Robots working in all their #warehouses. Each picking robot in the warehouse can replace 24 #workers #amazon #Robotics #humanoid @mvollmer1 @vinod1975 @helene_wpli @KirkDBorne @jeffkagan https://t.co/ylFLhlV6d7
Augmented reality + generative AGI means people can suddenly do expert-level work they never trained #AR #MR #XR #VR #GenAGI #GenAI #AugmentedReality #AI @anand_narang @antgrasso @Khulood_Almani @AlbertoEMachado @baski_LA https://t.co/dFrCsWVSh0

Ever tried using ChatGPT or Gemini for research, only to hit a wall when you need real multi-step analysis or technical synthesis? That’s why I tried OpenAI’s $200/Month plan for their research agent! After wasting hours piecing together reports and technical...

AI is transforming job roles by letting humans focus on creativity and strategy while automating routine tasks. Ready to shift from creator to curator? Learn more: https://t.co/99Xzl56S7R #AI #FutureOfWork #IoT #IBM https://t.co/HGXLy1QbNZ

OpenAI is upping the ante massively https://t.co/MmMAdIF7GL via @ft https://t.co/olx1X0v4DJ
What is @1x_tech up to later today at 11 a.m.? Launching its Neo home robot. Here's @karpathy who I surprised with my Apple Vision Pro on right after he hugged Neo while teleoperating it in one of the demos at its party...
Agreed. @flowith_ai's new OS is pretty mind blowing. More later after I play with it more. Just got home from @1x_tech which is launching its home-focused robot tomorrow. I'll have a lot more to say about that shortly. I am...
I just interviewed @karpathy for the first time in my life at @1x_tech. Right after he got a demo. I will have his video up later but turn up the audio and listen to what I have learned. The event here is...
Voices : How Much AI Is Too Much AI? https://t.co/GlQ91uudUG #AI #artificial_intelligence @ipfconline1 @JagersbergKnut @Khulood_Almani @Nicochan33
China leads global market as robot dogs become major AI testing ground | World News - Business Standard https://t.co/TBGLUzalZ3 #Robotics #robotdog #Robot #AI @CurieuxExplorer @Eli_Krumova @Enezator @enilev
I'm delaying sharing the videos of @blevlabs until tomorrow. They are long. In my house. An intimate look at 20 year old AI genius who has a system that I've seen over and over completely eviscerate results from multi-billion-dollar AI companies....
“Sora for Apps” For twenty years, software optimized what we clicked. Now it’s learning what we meant. We’re moving from tools that wait for instruction to systems that understand intent. The shift runs deeper than product design. It rewires the feedback loop...
What kind of leadership will define the future of energy? As we enter a decade where complexity is the new constant, I believe it’s on us, the global community of thinkers, builders, and leaders, to step up with vision, clarity, and...
How might the new X algo work? Not the shitty one we have today that's based on a bunch of hard-coded rules, but a more flexible AI-centric one? @bendiken demonstrated the other night a new kind of neurosymbolic machine intelligence. That builds...

🚨 I AM HIRING. Growth Marketer and 100% remote. You’ll own email and funnels for a 2M+ audience in AI advising and education. This is a rare opportunity - you don’t want to miss it. If you live for tests, journeys, conversion, apply...

Broadcom has established an entire knowledge workflow to maintain agent intelligence at scale. With this knowledge team process and a few 3P products in place, 88% of Broadcom's IT tickets are solved in under one minute. I think other businesses can...

some thoughts on @poolsideai's project horizon announcement last week: building a 2gw ai factory "This level of vertical integration - from model and agent development through to power generation and deployment - gives poolside a durable advantage as AI systems become...

If your friends or coworkers are still struggling with prompting or figuring out what AI can do for them, just tell them to type: Now what. Simple, powerful, easy to use, immediate ROI. Claude 4.5 and GPT-5 are powerful enough reasoning models to...
Today we shared an update about recent improvements to how ChatGPT responds in sensitive moments. We also published an update to the Model Spec to make some of our longstanding safety goals explicit. This is ongoing work and we’ll continue...

🎯 Scaling Ascent Peak: The Seven Summits of Artificial Intelligence In my latest article, I chart the evolution of AI—from the early days of symbolic logic to today’s autonomous agents, and what lies beyond. This journey is more than a history...
"Instant, interactive AI video." Boom!
FROM THE LIVE SHOW: “Even if a robot can cook better than Thomas Keller, you’ll still want to eat at the French Laundry, because you know Thomas Keller made the meal.” As AI floods the web with generated content, Runway (@runwayco)...
Good morning. Totally agree with @YunTaTsai1 below where he says robotics researchers shouldn't call themselves experts if they haven't had direct experience with @Tesla's FSD 14. In fact I'd go further. You can't call yourself a technology expert if you...
For a decade, creation meant learning the tools. Now the tools are beginning to learn from the creators. As @billpeeb, who leads Sora at OpenAI, put it: “Seventy percent of our users are creating content even to this day, a week and a...
It was such a pleasure to spend a Saturday morning with @RecursionChris recording this pod in NYC to talk about the @stateofaireport and biotech in 2025🦎 👀

🚨 From Agent-1 to Superintelligence: The AI 2027 Scenario The AI 2027 Report outlines one of the most thought-provoking trajectories for artificial intelligence: a rapid evolution from simple assistants (Agent-1) to autonomous, adversarially misaligned systems (Agent-4) and ultimately to Agent-5 —...
AI-powered soldiers: Are humanoid robots the future of warfare? https://t.co/jN4Kskwjd2 #warfare #humanoidrobotics #Robot #Robotics #humanoid @AlbertoEMachado @Eli_Krumova @postoff25 @Khulood_Almani @anand_narang

new zurich and munich events added to the @airstreet ai meetup calendar for 2026! stopping in a city near you soon - subscribe at luma(dot)com/airstreet https://t.co/2JYV1TVNOt
Meet WormGPT: The Chatbot That Doesn’t Say “No” to Hackers! https://t.co/YEKDzS6nvS via @The420in #WormGPT #GPT #hacker #chatbots @RagusoSergio @Shi4Tech @SpirosMargaris @vinod1975