Primer on AI and Machine Learning (Part 1) — Beginners Level (Non-Technical)
April 2022
Artificial Intelligence (AI) will be one of the greatest developments of science and civilization, as machines approach, augment, and exceed human performance on a wide range of cognitive tasks over the next few decades.
How should a smart, college-educated person with a fascination with AI but no technical background get up to speed on it? Below are the resources that I would suggest (I update this post about every 4 months). I offer five strategies to jump in:
Read some introductory articles;
Pick a few books and delve deeper;
Learn about AI/ML ethics;
Take a short course for non-technical people;
Follow the newsletters that best report on AI/ML.
I also have another post on AI/ML for more technical people (math and CS backgrounds) which is Part 2.
1) Articles to Get Acquainted with AI/ML
Non-tech overview of AI by Stanford panel: Start with this AI Index — this is the best overview from a panel of Stanford experts. The most recent report was “Gathering Strength, Gathering Storms: The One Hundred Year Study on Artificial Intelligence (AI100) 2021 Study Panel Report”.
McKinsey on the Impact of AI: One of the world’s best consulting firms gives you its take on where AI will go.
Bain on automation/AI: McKinsey’s biggest competitor has a different angle, focused on automation — basically AI is the biggest automatic force humans will ever see.
Economist Special Report on AI in business: This is a nice report looking at more specific applications on how AI will affect business.
The Economics of AI — a report from the new conference of economists researching this field.
Vertical AI Startups: This VC shows how much room exists for AI startups to add value, though I think larger companies will be able to do more first (because of their datasets).
HBR Article on Machine Learning (for business): This article by AI pioneer and professor Andrew Ng gives some tips for how businesses should ease into AI.
Machine Learning for Everyone: A simple, accessible explanation of ML.
Deep Learning Nature paper (not easy): This is a more technical and science-heavy paper about the hottest sub-field of machine learning, called deep learning. Warning — it’s a bit technical (it took me 4 months of study to really deeply understand ReLUs and fully-connected networks).
How Machine Learning Will Affect Jobs (CMU): One decent analysis on what jobs will be affected by AI versus what won’t.
The future of employment: how jobs are susceptible to computerisation: This Oxford report predicts 47% of US employment is at risk over 30 years, and I think that’s roughly right.
The wrong kind of AI? Artificial intelligence and the
future of labour demand. A brilliant research article from Daron Acemoglu at MIT arguing the current incentives are for companies to make AI that automates work instead of augmenting it.AI Hype and the Seven Deadly Sins (Rodney Brooks at MIT): A good takedown of AI hype, and focus on what is real versus what is not.
The Myth of Superhuman AI — AI is not about beating humans in everything, but augmenting and differentiating: A masterful article debunking the view of a single intelligence by Kevin Kelly, WIRED founder and seer — also states that AI will be as much about augmentation as automation.
Why general artificial intelligence will not be realized: I disagree with the conclusion of this paper, but it lays out the description of AGI (strong AI) better than most sources. IBM also has a page on Strong AI.
Early Warning Signals for AI — The Canaries: There’s too much hype about AI destroying the world, when in fact, the lack of AI causes car crashes, medical errors, wasted energy, and much suffering today. Here Oren Etzioni of the Allen Institute tells us when should take AGI Safety more seriously.
AI and the New Physics of Financial Services: An in-depth report from the World Economic Forum on how AI is drastically reshaping banking and finance.
“Rethinking Business Strategy in the Age of AI”: By HBS professors Iansiti and Lakhani, this is a solid overview of how APIs and AI will change business strategy.
Some Resources on Product Management for AI: This O’Reilly article has a good start. For a detailed take, look at these Digital Product Management notes (and section 5 for AI/ML specific notes).
2) Topical Books on AI/ML
Artificial Intelligence: A Guide for Thinking Humans: A best starting point and intro book to AI with some decent high-level propositions (though it’s weaker on predicting the future).
The Age of AI: And Our Human Future: A short and very high-level (easy) book on how AI will change the world, along with the challenges it will bring - a quick read.
Machine Learning Essential Knowledge (MIT): A short, readable primer on some ML basics. The book on Recommendation Engines is quite good too. Read it with Kelleher’s “Deep Learning (MIT Series)” book.
The Second Machine Age: Work, Progress, and Prosperity in a Time of Brilliant Technologies: A still relevant 2014 book by Erik Brynjolfsson and Andrew McAfee on AI and automation will change everything.
Architects of Intelligence: The truth about AI from the people building it: Martin Ford interviews the top experts in machine learning to hear how they think about the field. The range for when these experts say we’ll get “human-level AI” is from 10 years to over 150 years, with about 40 years being the median. A suprisingly good book.
Genius Makers: The Mavericks Who Brought AI to Google, Facebook, and the World: A decent recent history of the branch of AI, machine learning, that has stunned the industry by being very useful after 2012. Read it with Kelleher’s “Deep Learning (MIT Series)” book. For even more history, read Nilsson’s history of AI “The Quest for Artificial Intelligence.”
Competing in the Age of AI: How AI and big data change business strategy and economic thinking, as automation and centralization start to take over much of business and the global economy.
Real World AI: A Practical Guide for Responsible Machine Learning: A more practical book for business people and product managers figuring out where to build ML systems.
The Master Algorithm: How the Quest for the Ultimate Learning Machine Will Remake Our World: This gives a nice tour of the different branches of machine learning and their rich histories — it is the rare book that novices and experts can both appreciate.
Human Compatible: Artificial Intelligence and the Problem of Control: UC Berkeley AI professor Stuart Russell opines on the great promises and challenges of AI. He literally wrote the textbook.
Big Data Revolution: An early book on what “big data” is, as the essential pre-requisite and foundation of AI and ML.
Deep Medicine: How Artificial Intelligence Can Make Healthcare Human Again: A nice take on how AI will transform medicine — we are only 2% in for all the major changes to come. Who wants free or dirt cheap 24/7 health care from virtual assistants that combine the knowledge of the best human doctors?
Autonomy: The Quest to Build the Driverless Car―And How It Will Reshape Our World: One of the other big use cases of AI - self-driving cars and trucks, then eventually drones, flying cars, and starships.
The Book of Why: The New Science of Cause and Effect: Turing-award winner and UCLA Professor Judea Pearl lays out his vision for where AI should go — deep into causality, the holy grail.
Artificial Intelligence: A Modern Approach 4th Edition: The best textbook on AI, taught at UC Berkeley and many colleges. You can read it with a high-school math background and ideally some basic knowledge of coding. But I warn you that many approaches here are historical dead ends, other than machine learning and some tender other branches (the authors recently cleared out many older approaches that were in the 3rd edition when they reissued the 4th).
AI and ML hardware and chips:
AI Accelerator series (for more detail)
Hardware for Machine Learning: This is generally a neglected area, but this UC Berkeley ML hardware course is the best starting point. If you want to see how important hardware and GPUs are, check out this AI and Compute article from OpenAI.
3) Ethics and Governance of AI/ML (for non-technical people)
The Ethical Algorithm: A book about the complex problems and tradeoffs that ML gives when designing algorithms to serve business and society.
AI Ethics (MIT Press): A short and high-level intro to the field, a good starting point.
Weapons of Math Destruction: How Big Data Increases Inequality and Threatens Democracy: A sober look at how AI could really hurt us all — not through killer robots, but apps and recommenders that discriminate in hurtful ways and make inequality worse.
Robot Ethics 2.0: From Autonomous Cars to Artificial Intelligence: A good series of articles getting to the guts of AI ethics.
Harms of AI: An insightful paper by the economist Daron Acemoglu. I don’t agree with all of his speculations, but the ones backed up with empirical data (automation of labor causing wage stagnation) are compelling.
The Direct and Indirect Effects of Automation on Employment: A Survey of the Recent Literature: The unconventional survey paper on how AI and automation may increase jobs.
Ethics of AI systems: One article about ethics and AI — the books above are much better.
Courses on AI/ML Ethics: The only courses I’m aware of are Stanford’s CS182: Ethics, Public Policy, and Technological Change course taught by Rob Reich et al., UC Berkeley’s Human Contexts and Ethics of Data, and this CMU course Truth, Justice, and Algorithms that covers selected theoretical topics at the interface of computer science and economics. Oh, Stanford recently launched CS 384: Ethical and Social Issues in Natural Language Processing — which covers some real-life topics I’ve encountered in NLP AI systems I’ve built. Finally, here is an Ethics and Governance of AI Reading List from the Berkman Klein Center at Harvard.
Principles from US Institutes, Companies, and Governments: Google came out with some early Responsible AI practices, and this was matched with the Asilomar AI Principles that a smart, multi-disciplinary group of researchers came up with. The US Department of Defense’s DIA came out with these AI Principles (for warfighting!). Finally, here is an Ethics & Algorithms Toolkit, a risk management framework for governments put together by some data science researchers.
European AI Principles: Europe tends to take a heavier hand with AI ethics codes. The EU has its official stand in the EU Ethics Guidelines for Trustworthy AI. They also have the less onerous and somewhat vague OECD AI Principles and the quite good and detailed WEF report on Empowering AI Leadership An Oversight Toolkit for Boards of Directors.
Chinese AI Ethics: Meanwhile, in China, the only country on par with the US in AI (arguably better for AI in production tools, worse for AI research), first there are the Beijing AI principles, developed by the Beijing Academy of Artificial Intelligence (BAAI), an organization backed by the Chinese Ministry of Science and Technology and the Beijing municipal government. The code was developed in collaboration with the most prominent and important technical organizations and tech companies working on AI in China, including Peking University, Tsinghua University, the Institute of Automation and Institute of Computing Technology within the Chinese Academy of Sciences, and the country’s big three tech firms: Baidu, Alibaba, and Tencent. Second, there’s the Ministry of Science and Technology of the People’s Republic of China (MOST) which established a National Governance Committee for the New Generation Artificial Intelligence and released the Governance principles for the new generation of artificial intelligence — Developing responsible artificial intelligence. More here on Ethical Principles and Governance Technology Development of AI in China.
4) Simple Online Courses to Dive Deeper into AI/ML (for non-technical people)
AI For Everyone (Andrew Ng — Deep Learning): Andrew Ng is the brilliant former Stanford AI professor, co-founder of Google Brain (its AI group), and Baidu’s AI group. This is his short course explaining AI for the masses.
Elements of AI: A decent course set up by the government of Finland (in English and other languages) to teach everyone, everywhere about AI. I wish every national and state government had courses set up like this, customized to the job and automation needs of their region.
DeepMind’s “New to AI” videos: A nice series of videos from one the best AI groups in the world (a Google division).
5) The Best AI/ML Newsletters & Podcasts
Jack Clark Import AI: Newsletter by the in-house journalist at the non-profit OpenAI.
MIT The Algorithm: An MIT Tech Review newsletter on mostly AI topics.
KD Nuggets News: The top data science website publishing on ML.
Deeplearning.ai’s The Batch: Andrew Ng’s educational company puts out his great newsletter.
The Gradient: An AI Newsletter from Stanford
HAI Newsletter: Some informative news from the HAI center at Stanford - a good mix of technical and policy news
HBR Tech and Innovation: A decent HBR newsletter.
Exponential View: A newsletter exploring how our societies and political economy will change under the force of rapidly accelerating technologies and other trends.
AI Weekly: Pretty solid weekly roundup.
Lex Fridman: Find the AI ones from Lex, though generally all his science and tech episodes are quite good.
Robot Brains: From Pieter Abeel at UC Berkeley - simply amazing.
TWIML AI: A very sensible set of interviews with AI practitioners.