Letters from Andrew Ng

Personal messages to the AI community

Close-up of Andrew Ng speaking at World Economic Forum, backdrop reads "World Economic Forum" in white text.
Letters

From AI Experiments to AI Products: Experiemental projects can reveal AI powered-opportunities, but building AI-powered products often requires redesigning workflows.

How can businesses go beyond using AI for incremental efficiency gains to create transformative impact?
People golfing near a modern data center, set against a backdrop of hills and power lines at sunset.
Letters

Data Centers Are More Easier on the Environment Than You Might Think: How environmentally friendly are AI data centers? If we're going to use AI, there’s no better way to do it.

Many people are fighting the growth of data centers because they could increase CO2 emissions, electricity prices, and water use.
A birthday card generator form shows fields filled with humorous data and a chat bubble indicating help needed.
Letters

Build with Andrew!: AI enables people who don’t know any programming to build their own web apps using AI. Learn how in just 30 minutes with DeepLearningAI’s course “Build with Andrew.”

We just launched a course that shows people who have never coded before, in less than 30 minutes, how to describe an idea for an app and build it using AI.
Andrew Ng is pictured writing in a notebook by a large window, with a garden and pool visible in the background.
Letters

How to Test for Artificial General Intelligence: AGI has become a term of hype, and the traditional Turing Test can’t reliably detect it. How can we evaluate claims of that someone has built artificial general intelligence? Here’s a better test.

Happy 2026! Will this be the year we finally achieve AGI? I’d like to propose a new version of the Turing Test, which I’ll call the Turing-AGI Test, to see if we’ve achieved this.
Hands knit a red sweater with white snowflakes, sitting in a warm room with a crackling fire and snowy view.
Letters

How to Gain New Skills and Sharpen Old Ones: Give yourself a gift this holiday season by taking courses, building projects, and (maybe) reading papers.

Another year of rapid AI advances has created more opportunities than ever for anyone — including those just entering the field — to build software.
A robot examines a graph showing a rising trend in "amazingness" from 2015 to 2025
Letters

Large Language Models Are General — But Not _That_ General: Current progress in AI is piecemeal and laborious. Unforeseen breakthroughs may drive future progress, but the trend of improvement is incremental.

As amazing as LLMs are, improving their knowledge today involves a more piecemeal process than is widely appreciated.
Screenshot showing Python code generating a playable Snake game and the modern Snake game interface running on the right side.
Letters

Build an Autonomous Agent Using This Simple Recipe!: The ability of large language models to carry out multiple steps autonomously makes it possible to build a capable agent in a few lines of code.

If you have not yet built an agentic workflow, I encourage you to try doing so, using the simple recipe I’ll share here!
Heatmap illustrates countries' beliefs on AI's potential in solving issues like poverty and climate change.
Letters

Why People Don’t Trust AI and What To Do About It: Recent surveys by Edelman and Pew Research show that Americans distrust AI. The AI community should take this seriously and work to regain public trust.

Separate reports by the publicity firm Edelman and Pew Research show that Americans, and more broadly large parts of Europe and the western world, do not trust AI and are not excited about it.
A robot holds a bubble wand, surrounded by bubbles and colorful trees, with a futuristic city skyline.
Letters

Understanding the AI Bubble — If There Is One

Is there an AI bubble? With the massive number of dollars going into AI infrastructure such as OpenAI’s $1.4 trillion plan and Nvidia briefly reaching a $5 trillion market cap, many have asked if speculation and hype have driven the values of AI investments above sustainable values.
A diverse crowd engages with speakers at AI Dev x NYC, highlighting discussions on AI's impact and future.
Letters

What We Learned at AI Dev x NYC 2025: DeepLearning.AI’s sold-out AI Dev x NYC 2025 conference revealed widespread optimism, excitement, and technical depth among AI developers.

I just got back from AI Dev x NYC, the AI developer conference where our community gathers for a day of coding, learning, and connecting.
A megaphone emits a colorful stream of 3D words spelling "Hype", symbolizing the AI hype discussed in the article.
Letters

Don't Believe The Hype!: AGI is not just around the corner. People who enter AI today have huge opportunities to contribute to the field.

I recently received an email titled “An 18-year-old’s dilemma: Too late to contribute to AI?” Its author, who gave me permission to share this, is preparing for college.
Robots extract colorful data streams from silo towers, highlighting data silos being broken.
Letters

Tear Down Data Silos!: Many software-as-a-service vendors aim to hold their customers' data in silos. Their customers would do well to open the silos so AI agents can use the data.

AI agents are getting better at looking at different types of data in businesses to spot patterns and create value. This is making data silos increasingly painful.
Announcing the DeepLearning.AI Pro Membership!
Letters

Announcing the DeepLearning.AI Pro Membership!: The time to build with AI is now! One membership gives you all DeepLearning.AI courses, labs, practice sessions, and certificates for completed courses.

Today I’m launching DeepLearning.AI Pro — the one membership that keeps you at the forefront of AI. Please join!
Robot bakes pizza at 1000 degrees for 5 hours, causing a fire, illustrating mistake in error analysis.
Letters

Improve Agentic Performance with Evals and Error Analysis, Part 2: Best practices for error analysis in agentic AI development, and how LLMs make them easier

In last week’s letter, I explained how effective agentic AI development needs a disciplined evals and error analysis process, and described an approach to performing evals.
A man at a computer says AI ordered pizza, while a delivery man outside holds a fruit basket, highlighting a mix-up.
Letters

Improve Agentic Performance with Evals and Error Analysis, Part 1: When AI agentic systems go astray, it’s tempting to shortcut evals and error analysis. But these processes cas lead to much faster progress.

Readers responded with both surprise and agreement last week when I wrote that the single biggest predictor of how rapidly a team makes progress building an AI agent lay in their ability to drive a disciplined process for evals...

Subscribe to The Batch

Stay updated with weekly AI News and Insights delivered to your inbox