AI News – Global

The conversation of the moment is focused on one topic: AI agents. Unlike traditional language models that simply respond to a prompt, autonomous agents can execute multi-step plans and perform complex tasks on your behalf. But what happens when millions of these agents are not just working for us, but transacting, negotiating, and delegating to one another?

Nenad Tomašev, Senior Staff Research Scientist at Google DeepMind, joins host Hannah Fry to discuss the theoretical framework of a future”agentic economy.” Together, they discuss the operational shift from single systems to a cooperative “society of specialists,” the psychological risk of human automation bias, and the complex cybersecurity landscape—from dynamic cloaking to agentic traps—required to keep distributed intelligence secure.

Timecodes:
00:00 Intro
1:07 Defining AI agents
4:44 Agentic exploration in science and research
15:46 Delegation between agents
22:46 Agentic security and traps
29:31 Building an agentic economy
33:22 Cognitive monoculture
36:29 Distributed intelligence

To read the research, search for:
Distributional AGI Safety, May 2026
Intelligent AI Delegation, February 2026
Virtual Agent Economies, September 2025

Learn more about our AGI control roadmap: https://deepmind.google/blog/securing-the-future-of-ai-agents/

___

Subscribe to our channel https://www.youtube.com/@googledeepmind
Find us on X https://x.com/GoogleDeepMind
Follow us on Instagram https://instagram.com/googledeepmind
Add us on Linkedin https://www.linkedin.com/company/deepmind/

The co-founders of Anthropic discuss the past, present, and future of Anthropic. From left to right: Chris Olah, Jack Clark, Daniela Amodei, Sam McCandlish, Tom Brown, Dario Amodei, and Jared Kaplan.

Links and further reading:
Anthropic’s Responsible Scaling Policy (RSP): https://www.anthropic.com/news/announcing-our-updated-responsible-scaling-policy
Machines of Loving Grace: https://darioamodei.com/machines-of-loving-grace
Work with us: https://anthropic.com/careers
Claude: https://claude.com

00:00 Why work on AI?
02:08 Scaling breakthroughs
03:30 Early days of AI
10:57 Sentiment shifting
18:30 The Responsible Scaling Policy
30:42 Founding story
32:45 Building a culture of trust
39:08 Racing to the top
43:43 Looking to the future

Bryant Chou co-founded Webflow, which today powers around 1% of all websites on the internet. Now he’s back in the current YC batch with Ploy, an AI-powered website and marketing platform that doesn’t just build your site — it connects to your analytics, CRM, and search console to optimize your marketing while you sleep. In this episode of the Lightcone he explains how he built Ploy to be “anti slop,” how building today compares to his first startup, and why founders with domain expertise are making a comeback.

Chapters:
00:00 — What Experience Gives You in the Age of AI
00:38 — Meet Bryant Chou, Co-Founder of Webflow
01:22 — His New Startup Ploy
02:47 — Rebuilding the Posterous website From 2008
03:27 — Rebuilding the Scribd website From 2007
05:04 — Rebuilding the Auctomatic website From 2007
06:19 — Rebuilding the Escher Reality website From 2017
07:11 — 12% of the YC Batch Uses Ploy
08:26 — The D&D Theory of Founder Skills
10:05 — Democratizing Marketing and Growth
10:50 — Live Demo: The Design Slurper
13:21 — Your Website Should Work for You While You Sleep
14:26 — Integrations, Analytics, and the Marketing Brain
17:27 — Ploy’s Anti-Slop Engine: 3,500 Curated Design Prompts
20:05 — The Andy Warhol Theory of AI
22:35 — Webflow Origin Story
24:26 — Building in a Competitive Market Then vs. Now
26:01 — First Three Months: Webflow 2013 vs. Ploy 2025
27:17 — What Experience Teaches You That Models Can’t
28:51 — Will Better Models Kill Products Like Ploy?
30:32 — The Competitive Moat of Purpose-Built AI
33:01 — Agents as Customers: CLI, MCP, and AEO
35:02 — Young Founders vs. Experienced Founders
36:36 — The Idea Maze and Cloning Yourself With AI
42:37 — The Magnifying Glass Moment

Apply to Y Combinator: https://www.ycombinator.com/apply
Work at a startup: https://www.ycombinator.com/jobs

Buy the AI Engineering book here to continue your learning! https://amzn.to/42kjXb2

All images are from the book AI Engineering unless otherwise credited.

⏰ Timestamps
00:00 What is AI Engineering?
01:49 Understanding Foundation Models
08:40 Evaluating AI Models
14:50 Model Selection
23:15 Prompt Engineering
30:20 RAG and Context Construction
36:56 Agents and Memory Systems
43:02 Finetuning
52:40 Dataset Engineering
59:45 Inference Optimization
01:09:01 Architecture and User Feedback
—————————————-
Here’s a list of my favorite free AI/ML courses! https://www.gratitudedriven.com/free-aiml-courses

🚀 Want to level up your AI/ML career? Join the AI/ML Career Launchpad community! https://aiml-career-launchpad.circle.so/aiml-launchpad

💬 Interested in working with me 1:1? Learn more about my strategic advisory sessions: https://www.marinawyss.com/coaching

🩷 Join the channel membership community for priority comment replies and early access to videos! https://www.youtube.com/@MarinaWyssAI/join

And here’s that StatQuest transformers video I promised:
—————————————-
🎥 Other videos you might like:

Doubting if You Can Become a Machine Learning Engineer or Data Scientist? The Honest Truth.

I Analyzed 1,238 Machine Learning Job Postings to Find Out What Skills You REALLY Need to Get Hired

Beginner to (Employed) Data Scientist in 2025: Complete Roadmap – Skills, Projects, CV, Interviews

—————————————-
🦫 About me
I am an Applied Scientist (basically, a blend of Data Scientist/Machine Learning Engineer) at Twitch/Amazon. Outside of my full-time job I’m a technical mentor at a machine learning bootcamp, and a 1:1 career coach for people looking to break into the field, with a focus on those from non-traditional backgrounds.

I’m also a Certified Personal Trainer, always busy with too many interests, and really, deeply happy with my life. I hope to be able to help others achieve these things, too.

Instagram: / gratitudedriven
—————————————-
✉️ Contact
Leave me a comment here on YouTube!
Business email: business@gratitudedriven.com
—————————————-
⚖️ Disclaimer
The views and opinions expressed in this video are my own and do not reflect the official policy or position of Twitch/Amazon or any other company I have worked for. All advice and insights shared here are based on my personal experiences and should be considered as such.

Also, there may be affiliate links in this description. This means if you make a purchase I’ll earn a small commission at no cost to you! Thank you for your support.

#aiengineering #ai #machinelearning

AI data centers, energy crisis, US vs China tech war, Saudi Arabia AI investment, cloud infrastructure & electricity grid, explained.
This episode exposes why data centers are the biggest bet in the AI race.

I cover:
00:00 Why everyone wants data centers
03:04 Why electricity is becoming the new oil, and why you can’t just make more of it
07:13 How the US grid is hitting a ceiling
11:01 Why China built 500 data centers and 80% sit empty
15:23 Why the Gulf – Saudi Arabia, UAE – is becoming the third player in the AI infrastructure race
19:21 Chinese and American ties with the GCC

🪙 This is Episode 1 of The Business of AI: a series tracing the hidden web of capital, geopolitics, and infrastructure underneath every AI headline. We’re starting at the bottom of the stack.

Every time you use ChatGPT, or Claude, or Gemini, money flows through a stack of cloud providers, GPUs, and data centers. This is a deep dive into the real economics behind AI subscription costs.

Chapters:
00:00 Intro
00:42 Who gets paid when you run an AI query
02:32 Why compute is SO expensive
07:07 The breakdown of your $20 subscription
07:35 How Azure, AWS, and CoreWeave profit on every level
11:11 Why data centers sell power, not space
12:19 Time for delusional startup ideas

🌟Create any visuals for your business with Higgsfield
https://higgsfield.ai/s/mcp-techbutmakeitreal-RZOKvq

Breakdown of the entire semiconductor supply chain, from TSMC and ASML to Japan’s chemical monopolies, Dutch lithography machines, America’s chip software, German optics, to explain how the global semiconductor industry actually works.

Chapters:
0:00 Intro
0:50 EDA Software: America’s Chip Design
1:58 Silicon Wafers: Japan’s Precision Monopoly
3:29 Photoresist: The Chemical Ink of Chips
7:05 Neon Gas and Ukraine’s Role
9:06 ASML: The Dutch Gatekeeper
10:19 Zeiss, Cymer, and US Control over the Dutch
11:33 Why China Can’t Replicate ASML
12:16 Etch, Deposition, and Inspection Equipment
14:15 Photomasks, Pellicles, and Lasertec
15:17: TSMC, Samsung, and the Foundry Race
18:20 Why Chip Fabs Cost More Than Skyscrapers
19:15 BIG Questions

#Semiconductors #TSMC #ASML #AI #Nvidia #China #Taiwan #Geopolitics #TechButMakeItReal #ChipWar #ArtificialIntelligence #Samsung #Intel #EUV #SemiconductorIndustry

The unit economics behind LLM compute, 2% paid users, and 30-60% margins.

This episode breaks down why many AI-native apps (GPT wrappers, agent tools, “AI SaaS”) struggle to become real businesses. I compare classic B2B SaaS (70–90% gross margins) with GenAI’s 30-60%, unpack ongoing per-user costs (API, compute, model licensing, moderation), and explain how price caps, freemium, and “dead subs” distort the picture.

Chapters
00:00 Intro
00:30 Classic SaaS vs AI SaaS margins
01:56 “Dead subscriptions” and AI bundles
03:55 Why AI unit economics break
05:48 All-you-can-eat usage
06:39 The shocking math of ChatGPT
09:02 Claude, Gemini, Copilot
12:15 AI price wars, metering, and profitability paths
13:26 How to build a sustainable AI business
15:12 $1,000,000,000 problem
16:52 Bottom line

The example-driven, practical walkthrough of Large Language Models and their growing list of related features, as a new entry to my general audience series on LLMs. In this more practical followup, I take you through the many ways I use LLMs in my own life.

Chapters
00:00:00 Intro into the growing LLM ecosystem
00:02:54 ChatGPT interaction under the hood
00:13:12 Basic LLM interactions examples
00:18:03 Be aware of the model you’re using, pricing tiers
00:22:54 Thinking models and when to use them
00:31:00 Tool use: internet search
00:42:04 Tool use: deep research
00:50:57 File uploads, adding documents to context
00:59:00 Tool use: python interpreter, messiness of the ecosystem
01:04:35 ChatGPT Advanced Data Analysis, figures, plots
01:09:00 Claude Artifacts, apps, diagrams
01:14:02 Cursor: Composer, writing code
01:22:28 Audio (Speech) Input/Output
01:27:37 Advanced Voice Mode aka true audio inside the model
01:37:09 NotebookLM, podcast generation
01:40:20 Image input, OCR
01:47:02 Image output, DALL-E, Ideogram, etc.
01:49:14 Video input, point and talk on app
01:52:23 Video output, Sora, Veo 2, etc etc.
01:53:29 ChatGPT memory, custom instructions
01:58:38 Custom GPTs
02:06:30 Summary

Links
– Tiktokenizer https://tiktokenizer.vercel.app/
– OpenAI’s ChatGPT https://chatgpt.com/
– Anthropic’s Claude https://claude.ai/
– Google’s Gemini https://gemini.google.com/
– xAI’s Grok https://grok.com/
– Perplexity https://www.perplexity.ai/
– Google’s NotebookLM https://notebooklm.google.com/
– Cursor https://www.cursor.com/
– Histories of Mysteries AI podcast on Spotify https://open.spotify.com/show/3K4LRyMCP44kBbiOziwJjb

– The visualization UI I was using in the video: https://excalidraw.com/
– The specific file of Excalidraw we built up: https://drive.google.com/file/d/1DN3LU3MbKI00udxoS-W5ckCHq99V0Uqs/view?usp=sharing
– Discord channel for Eureka Labs and this video: https://discord.gg/3zy8kqD9Cp

Educational Use Licensing
This video is freely available for educational and internal training purposes. Educators, students, schools, universities, nonprofit institutions, businesses, and individual learners may use this content freely for lessons, courses, internal training, and learning activities, provided they do not engage in commercial resale, redistribution, external commercial use, or modify content to misrepresent its intent.