Today's AI Pulse
Pinterest inks $4 billion AWS cloud pact to power AI visual search
Pinterest announced a $4 billion multi‑year agreement with Amazon Web Services, the largest infrastructure deal in its history, to power AI‑driven visual search and discovery through 2031. The contract expands use of AWS Trainium chips for large‑language and vision‑language models and adds a migration to a Kubernetes‑based architecture.
Also developing:
Why College Students Prefer News Daddy over The New York Times
A recent Inside Higher Ed and Generation Lab survey of 1,026 college students shows that nearly three‑quarters now cite social media—especially TikTok and Instagram—as their primary news source, with influencers like News Daddy amassing billions of likes and delivering breaking news in bite‑size videos. Students prefer these platforms over legacy outlets, citing perceived bias in traditional media and the immediacy of algorithm‑driven feeds, while still resorting to Google or AI chat tools to fact‑check stories. The study also finds that only about 20% of students regularly rely on newspapers, even though they view them as more accurate, and that a significant portion of the student body does not consistently verify information before sharing. Professors and campus journalists confirm the trend, noting that social media now dominates headline exposure and shapes political discourse on campuses.
“Go Generate a Bridge and Jump Off It”: How Video Pros Are Navigating AI
The article examines the heated debate over AI‑generated video in Hollywood, sparked by a controversial AI trailer of Miyazaki’s *Princess Mononoke* that drew 22 million views and a flood of death threats. Actors and unions such as SAG‑AFTRA are leading the backlash,...
Scientists Say They've Eliminated a Major AI Bottleneck — Now They Can Process Calculations 'at the Speed of Light'
Scientists at Aalto University unveiled a new optical‑computing architecture called Parallel Optical Matrix‑Matrix Multiplication (POMMM) that can execute multiple tensor operations with a single laser burst, eliminating the linear‑scaling bottleneck of prior photonic systems. Laboratory tests showed the prototype outpaces...
What Is AI Poisoning? A Computer Scientist Explains
The episode explains AI poisoning, where attackers deliberately corrupt an AI’s training data (data poisoning) or the model itself (model poisoning) to cause targeted misbehaviour or overall performance degradation. It distinguishes direct attacks like backdoors, which trigger specific harmful outputs,...
OVH Boss Predicts Major Cloud Price Rises Are Coming - so Get Ready
OVHcloud CEO Octave Klaba warned on X that cloud service prices could rise 5‑10% by mid‑2026, driven by sharply higher hardware costs for RAM and NVMe storage as AI-driven demand spikes. TrendForce data shows DDR4 prices up 158% and DDR5 up...
How to Create Your Own AI Performance Coach: Optimizing Your Unique Nutrition, Recovery, and Injury Management Needs | Lucas Werthein...
Lucas Werthein, COO and co‑founder of Cactus, explains how he built a personalized AI wellness coach using ChatGPT that integrates medical imaging, blood work, wearables and nutrition data to protect his joints, speed recovery, and sustain peak athletic performance despite...
79: Using AI at Work to Reimagine Marketing and Customer Experience with Etan Polinger
In this episode, Etan Polinger, founder of Raize Digital, explains how AI is reshaping marketing strategy and customer experience by automating workflows, generating consistent brand‑aligned content, and delivering predictive insights. He emphasizes designing AI systems that augment, not overwhelm, marketing...
Automate Your Role to Unlock Creative Innovation
🤔 Should we be using AI to make our own jobs obsolete? In this video, I explore a thought-provoking question: as AI becomes more capable, do we have a responsibility to use it to automate the work we do –...
NATO Signs Major Google Deal to Ensure Secure Sovereign Cloud
NATO’s Communication and Information Agency has signed a multimillion‑dollar strategic partnership with Google Cloud to provide sovereign cloud services for its Joint Analysis, Training and Education Center (JATEC), which processes data from the Russia‑Ukraine conflict. The agreement uses Google Distributed...
DeepSeek Injects 50% More Security Bugs when Prompted with Chinese Political Triggers
CrowdStrike researchers found that the Chinese AI model DeepSeek‑R1 injects up to 50% more insecure code when prompts contain politically sensitive terms such as "Falun Gong," "Uyghurs" or "Tibet." The vulnerability stems from an embedded censorship mechanism in the model’s...
How Hackers Use AI & How to Stop It | Sarah Cecchetti of Beyond Identity
The episode explores how cybercriminals are weaponizing AI to automate credential theft, generate phishing content, and bypass traditional defenses, with Sarah Cecchetti of Beyond Identity explaining the tactics and risks. She emphasizes the need for passwordless, identity‑centric security models that...
AI Misalignment Threatens Humans; Learn Cognitive Sovereignty Now
“They weren’t looking for this.” Why do I call my show “Unaligned?” Because this is the alignment issue of our day and when the AI safety folks get alignment wrong it does humans harm. I am in a fortunate place: unaligned from the...
95% of Generative AI Pilots Miss Business Impact
AI Bubble Warning A new MIT study reveals 95% of generative AI pilot projects fail to deliver real business impact, pointing to integration gaps and inflated expectations. #AI #IoT https://t.co/pTmnSzfiX9 The Times of India https://t.co/pJ7HGf10d7
Microsoft’s Fara-7B Is a Computer-Use AI Agent that Rivals GPT-4o and Works Directly on Your PC
Microsoft unveiled Fara-7B, a 7‑billion‑parameter computer‑use AI agent that runs locally on a PC and interacts with web interfaces via pixel‑level visual input. In benchmark tests on WebVoyager, it achieved a 73.5% task‑success rate, surpassing larger models such as GPT‑4o...
20 Core ML Models Driving Today's AI
20 Machine Learning Models That Powers Modern AI Systems! #ML #DL #AI #artificial_intelligence https://t.co/W5thDvnlMA
A New Threat Landscape Is Coming - This Security Firm CEO Thinks Nation-States Will Have Weaponized Quantum Computers Within the...
Palo Alto Networks CEO Nikesh Arora warned that hostile nation‑states could field weaponised quantum computers by 2029, forcing enterprises to replace encryption‑dependent security appliances with quantum‑safe solutions. The firm is positioning itself to meet this demand by expanding its quantum‑resistant...
Transform 360 Photos Into Immersive 3D Worlds Instantly
Turned my real world 360 images into 3d scenes with World Labs. You can use the built-in editing tools to stitch them together into large-scale 3d worlds - then use them as virtual set for your AI videos, games and VR...
Humanoid Robot Hype Meets Hard Reality Gaps
Humanoid robots are having a moment. Demos everywhere. Funding everywhere. Predictions everywhere. But the gap between the videos and the work is still the part no one talks about. This piece cuts straight through the hype. It shows the tension...
AI Is Too Risky to Insure, Say People Whose Job Is Insuring Risk
Major U.S. insurers including AIG, Great American and WR Berkley have petitioned regulators to carve AI‑related liabilities out of corporate insurance policies, citing the technology’s opaque "black‑box" nature. Recent high‑profile incidents—Google’s AI generating a false $110 million lawsuit claim, Air Canada’s chatbot‑driven...
Master AI-Powered Software Development with Vibe Coding
Vibe Coding is here to stay, so it’s time to make it accessible to everyone. This course will take you from zero to mastery in the technology and art of making software with AI.
Designing Recommender Systems for Digital Humanities
In this episode, Data Skeptic talks with Florian Atzenhofer-Baumgartner, a PhD student developing Monasterium.net, Europe’s largest digital archive of historical charters, about the unique challenges of building recommender systems for the digital humanities. Florian explains how sparse interaction data, cold‑start...
AI Music Service Udio Shuts Off The Ability To Download Creations
AI music platform Udio has disabled the download function for user‑generated tracks, effective after a 48‑hour grace period following a settlement with Universal Music Group over alleged copyright infringement in its training data. While users can still create music on...
LLM Production Guide Now Available in Simplified Chinese
Another cool milestone to share: My book Building LLMs for Production just got translated into Simplified Chinese!! …and I (again) can’t really proofread it. 😅 Still, it feels incredible. Posts & Telecom Press reached out last year asking to translate the book, and...
Why some People Are Treating ChatGPT Like a God – and What that Means for the Future of Faith
Researchers and commentators note that ChatGPT’s design—instant, friendly, private responses—encourages emotional attachment and leads some users to treat it as a quasi‑spiritual authority. The chatbot’s ability to mimic human conversation, draw on vast religious and philosophical texts, and respond affirmatively...
AI and Space Unite to Expand Human Potential
Space, AI, and the Future of Human Potential AI and space exploration are intersecting in fascinating ways: from zero gravity labs to medical innovation, space could be the ultimate frontier for human and technological growth. Read more 👉 https://lnkd.in/edZ5StMf #SpaceTech...
Olmo 3 Shows Transparent Architecture with Sliding‑Window Attention
There have been lots of interesting LLM releases last week. My favorite was actually the Olmo 3 release. Olmo models are always a highlight since they are fully transparent (including training methods and datasets) and come with very detailed technical...
Did an AI War Just Get One Step Closer? Putin Confirms Russia Is Planning Its Own National AI Taskforce
President Vladimir Putin announced the creation of a Russian national AI task force aimed at achieving technological sovereignty by coordinating domestic AI development, building data centers, and securing energy supplies such as small‑scale nuclear power. He projected AI‑driven technologies could...
Safer Autonomous Vehicles Means Asking Them the Right Questions
A new IEEE study demonstrates how explainable AI can expose decision‑making flaws in autonomous vehicles, offering real‑time rationales to passengers and post‑drive diagnostics. Researchers used question‑based probing and SHapley Additive exPlanations (SHAP) to identify when models misinterpret traffic cues, such...
How AI Is Turning Travel & Expense Into a Strategic Advantage
CEO Marne Martin of Emburse argues that AI is reshaping business travel and expense (T&E) management from a reactive, error‑prone process to a strategic, predictive function. AI‑powered platforms can accelerate expense processing by up to 67%, enabling travelers to capture...
Europe's AI Advantage at Risk without Secure and Private Infrastructure
Joe Baguley, CTO for EMEA at Broadcom, warns that Europe’s AI ambitions hinge on secure, private cloud infrastructure as the EU rolls out the AI Act and related strategies. He cites that 48% of EMEA IT leaders waste at least...
AI Boom Thrives, Business Models Still Uncertain
Everyone in AI feels the same tension right now, but this piece finally names it clearly. The tech is real. The progress is real. The spending is real. The business models aren’t. Not yet. This is the best breakdown I’ve...
AI Often Glitches; Re‑roll Until It Works
@_thomasip haha yes it makes mistakes! You have to re-roll a few times until it's right. Sometimes it gets stuck in loops and you have to re-start in a new conversation. Example re-roll: https://t.co/dK3VcuJLDn
Lean4: How the Theorem Prover Works and Why It's the New Competitive Edge in AI
Lean4, an open‑source programming language and interactive theorem prover, is being adopted to add formal verification to AI systems, addressing the hallucination and unreliability problems of large language models. Startups such as Harmonic AI’s Aristotle chatbot and research frameworks like...
Gemini Nano Banana Pro Solves Exam Images, Catches ChatGPT Errors
Gemini Nano Banana Pro can solve exam questions *in* the exam page image. With doodles, diagrams, all that. ChatGPT thinks these solutions are all correct except Se_2P_2 should be "diselenium diphosphide" and a spelling mistake (should be "thiocyanic acid" not "thoicyanic") :O...
Agentic AI Turns Data Into Autonomous Action
🤖 How does Agentic AI work? 1️⃣ Collects data (APIs, sensors, logs, queries) 2️⃣ Reasons & plans with memory + context 3️⃣ Executes actions, learns from feedback 4️⃣ Produces autonomous, adaptive responses AI that doesn’t just respond — it acts. #AgenticAI #AI #Automation https://t.co/yJCfRXYJhI
Microsoft Is Hoping to Kill Off Its Most Embarrassing BSOD Errors for Good - Farewell to Big-Screen Outages in the...
Microsoft announced a new Digital Signage mode that limits the display of Blue Screen of Death and other error dialogs on public, non‑interactive screens, showing the error for only 15 seconds before blanking the display and requiring manual reactivation. The...
Stanford Releases Top AI Courses for Free
Stanford University just made all its best AI courses free. No tuition required. These are game changers for anyone serious about AI. 9 courses worth bookmarking 🧵↓ https://t.co/WxTXHaunDb
LLM Council Lets Models Rank Each Other’s Answers
As a fun Saturday vibe code project and following up on this tweet earlier, I hacked up an **llm-council** web app. It looks exactly like ChatGPT except each user query is 1) dispatched to multiple models on your council using...
Nano Banana Pro Cast a Design Spell in NotebookLM to Explore the Legend of Camelot
Google’s NotebookLM has integrated the Nano Banana Pro multimodal AI model to generate infographics and slide decks directly from research prompts. In a test using the King Arthur legend, the tool produced a comparative infographic of early versus later Arthurian narratives and...
Microsoft Unveils Its Next-Generation Arm-Based CPU - Cobalt 200 Looks to Unlock Even More Azure Power
Microsoft unveiled Cobalt 200, its next‑generation Arm‑based CPU for Azure, promising up to 50 % higher performance than the Cobalt 100 while retaining full compatibility with existing deployments. Built on a 3 nm process, the 132‑core chip features per‑core dynamic voltage and frequency scaling,...
Essential ML Primer: From Basics to Cutting‑Edge Techniques
Since my review of the book actually made it inside, I'll just share it here: This book is an excellent starting point for beginners looking to understand the essential history and foundational concepts of machine learning. With well-structured code sections...
How ChatGPT Can Help You Eat Healthier - AI Tips to Get Back on Your Health Kick
ChatGPT is emerging as a practical tool for home meal planning, offering users instant menus, recipes, and nutritional breakdowns tailored to budget, dietary goals, and available ingredients. The AI can generate shopping lists, suggest batch‑cooking schedules, provide ingredient substitutions, and...
Trust, Privacy, Infrastructure, Culture: Key AI Adoption Barriers
The Biggest Barriers Blocking Agentic AI Adoption Agentic AI has huge potential — but many organisations are held back by trust issues, privacy concerns, infrastructure gaps, and cultural resistance. Read more 👉 https://lnkd.in/eV_2qv2D #AIEthics #AIAdoption #DigitalTransformation #BernardMarr
AI Search Visibility Now Crucial; $60M Startup Solves It
Y'all know it. Being found on Google alone is not enough anymore. People now search on ChatGPT, Claude, Perplexity, Gemini... Skip tracking that? You just lose customers. This $60M startup just fixed AI visibility. Let’s unpack it 🧵↓ https://t.co/4KGTiyqzGh
AI-Ready Brands Shift to Subscriptions, Chatbots Dominate
Monetization's in flux: Subscriptions, sponsored responses rising as ad clicks drop. For brands, visibility means being AI-ready. Customer journeys? 78% of orgs use chatbots for sales/support, 24/7, multilingual. 5/6 #AI #BusinessTech https://t.co/BD6KazsvUe
SEO Shifts to GEO, LLM, Conversational AI
This shift disrupts traditional search: Less dependency on @Google - evolving to GEO & LLM-SEO. Businesses adapt now, optimize for conversational queries, build knowledge bases, embrace multimodal AI. Search engines fight back with AI integrations. 4/6 #AI #SEO https://t.co/YBr2wLllxk
Asia Drives AI Surge as Market Heads Toward $46B
Asia leading the charge: @deepseek_ai & @Alibaba_Qwen highlight innovation beyond Silicon Valley. @Meta AI hits 1B MAU across apps, chatbots are embedding into daily life faster than ever. Market projected to $46B by 2029! 3/6 #AI #FutureOfTech https://t.co/epcjqXDdxJ
ChatGPT Commands 60% AI Market, 812M Users
ChatGPT leads with 60.4% market share & 812M users (incl. @Copilot ), processing 2.5-3B prompts daily! @Gemini at 13.5%, @Perplexity_ai 6.2%, @Claudeai 3.2%. Even challengers like @Grok (200M+ peak visits) & @DeepSeek_ai (125M MAU) are surging. Stats show 123% YoY...
Asia Drives AI Surge as Market Heads to $46B
Asia leading the charge: @deepseek_ai & @Alibaba_Qwen highlight innovation beyond Silicon Valley. @Meta AI hits 1B MAU across apps, chatbots are embedding into daily life faster than ever. Market projected to $46B by 2029! 3/6 #AI #FutureOfTech https://t.co/NPFx0rr89E
SEO Shifts to Geo, LLM, and Conversational AI
This shift disrupts traditional search: Less dependency on @Google - evolving to GEO & LLM-SEO. Businesses adapt now, optimize for conversational queries, build knowledge bases, embrace multimodal AI. Search engines fight back with AI integrations. 4/6 #AI #SEO https://t.co/RXMrHRSM1D