Michael I. Jordan on Machine Learning as Engineering, the Limits of AI, and the Future of Decision-Making Systems

Michael I. Jordan, a leading figure in machine learning, argues that the current state of "AI" is not about achieving human-level intelligence but rather the emergence of a new engineering discipline, akin to chemical or electrical engineering. He posits that this field, which he prefers to call machine learning or data science, focuses on building viable, safe, and scalable systems that bring value to human beings by making large collections of decisions under uncertainty, often involving human data and decisions. He strongly distinguishes this from the philosophical aspiration of true artificial intelligence, which he believes is centuries away due to our profound lack of understanding of the human brain. Jordan draws a sharp line between current AI capabilities, which he largely categorizes as advanced engineering and pattern recognition, and genuine scientific understanding of intelligence or the brain. He criticizes the "overselling" of AI breakthroughs, particularly demos that mimic human behavior (like a computer calling a restaurant), as mere stunts rather than fundamental scientific progress. He contrasts his view with those who prioritize prediction (like Yann LeCun), arguing that decision-making, especially consequential decisions in the real world with real-world risks, economic implications, and human impact, is equally, if not more, crucial and often overlooked. The discussion highlights the need for a robust engineering field that integrates statistical and computational ideas to build systems that operate at planetary scale, considering economic viability, safety, and externalities like privacy. Jordan uses the example of a "music market" to illustrate a missing market where creators (musicians) are not directly compensated or empowered by the data flows their work generates, suggesting that current systems (like Spotify) primarily benefit intermediaries. This points to a need for transparent, vetable systems that empower creators and consumers through direct economic relationships based on data. The conversation underscores the ethical and societal implications of deploying AI systems at scale, particularly concerning privacy, economic fairness, and the responsible management of consequential decisions. Jordan's perspective calls for a more sober, principled approach to developing these technologies, emphasizing scientific understanding and predictable, valuable outcomes over exuberant promises. It suggests that the true breakthroughs of this era will be in establishing robust, principled engineering disciplines that can manage complex, distributed decision-making systems, much like previous centuries saw the rise of electrical and chemical engineering.

I think what happening right now is not ai that was an intellectual aspiration that's still alive today is an aspiration but I think this is akin to the development of chemical engineering from chemistry or electrical engineering from from electromagnetism.

we have no clue how the brain does computation we're just a clueless we're like we're even worse than the greeks almost anything interesting scientifically of our era

it is the problem of the next few centuries it is fantastic but we have our metaphors about it is it an economic device is it like the immune system or is it like a layered you know set of copy you know arithmetic computations what we have all these metaphors and they're fun but that's not real science per se

I don't think that having a computer call a restaurant and pretend to be a human is a breakthrough and people you know some people present it as such it's imitating human intelligence it's even putting coughs in the thing to make a bit of a pr stunt

I don't need to be wowed but I I think that someone comes along in 20 years a younger person who's absorbed all the uh the technology and for them to be wowed I think they have to be more deeply impressed a young kulmogorov would not be wowed by some of the stunts that you see right now coming from the big companies the demos

trying to program a computer to understand natural language to be involved in a dialogue we're having right now that can happen in our lifetime you could fake it you can mimic sort of take old sentences that humans use and retread them with the deep understanding of language now it's not going to happen

I don't want to reclaim it I want a new word I think it was a bad choice I mean I I you know I if you read one of my little things um the history was basically that uh mccarthy needed a new name because cybernetics already existed and he didn't like you know no one really liked norbert wiener

I'm not going to trust the output of that neural net to predict my heart attack I'm going to want to ask what if questions around that I'm going to want to look at some us or other possible data I didn't have causal things I'm going to have a dialogue with a doctor about things we didn't think about we gathered the data you know it I could go on and on I hope you can see and I don't I think that if you say predictions everything that that you're missing all of this stuff

the goal of that is really good working systems at planetary scale we've never seen before

there is no music market in the world right now or in the con in our country for sure there are something called things called record companies and they make money and they prop up a few um really good musicians and make them superstars and they all make huge amounts of money but there's a long tale of huge numbers of people that make lots and lots of really good music that is actually listened to by more people than the famous people they are not in a market they cannot have a career they do not make money