Every year, MIT Technology Review releases its list of 10 Breakthrough Technologies , highlighting the innovations most likely to shape how we live, work and tackle the world’s biggest challenges. These technologies span a range of fields and offer more than a fascinating glimpse into the future — they reveal the technologies already shaping the present in profound ways.
Amy Nordrum, executive editor of MIT Technology Review, joined GBH’s All Things Considered host Arun Rath to walk through the 2025 picks. What follows is a lightly edited transcript.
Arun Rath: Before we go through this year’s picks, I want to get a better understanding of what goes into compiling this amazing list. As the executive editor, what are the criteria you use to curate the list and determine whether a technology is a real breakthrough?
Amy Nordrum: Sure. Well, our newsroom covers emerging technology, and every year, our journalists and editors get together and talk about which technologies we think are most likely to have the greatest impact on our future, the way we live and the way we work.
That can mean a couple of different things. We look for scale with a lot of these technologies, so something that is going to have very wide adoption or maybe is going to remake a very important industry that we all rely on in our daily lives. Or, it could be a very transformational scientific breakthrough that’s really changing the game in a certain area of medicine that has a lot of patient demand.
So, it can mean a couple of different things depending on the field that we’re looking at, but what we want is something that’s really high-impact and also something that’s kind of at a breakthrough moment. There’s a reason — or a tipping point, if you will — why we think, this year, this technology has reached a point where it’s more likely than not to be widely adopted or to be able to commercially scale in the future.
Rath: Well, let’s start in the area of robotics. Tell us about the category and how you think what you’re writing about could transform industries or everyday life.
Nordrum: Right. We have a couple of robotics items on the list this year.
I’ll start with fast-learning robots. We picked this one because it’s exciting to see some new advances in artificial intelligence, in particular, really accelerate progress in other fields. So, for robotics, this is a field that has for a long time kind of struggled to make robots that can do really useful stuff.
I’ve personally seen in my career as a tech journalist a lot of robot demos, and they often will show robots doing really specific tasks in a very controlled lab setting. But we’ve yet to see a robot that can do all kinds of stuff, different tasks very easily in a real environment that’s changing all the time.
But now, we are getting closer because of some new data sets and training techniques that roboticists have developed, many of which make use of generative AI and large language models.
Rath: Wow. So, the next generation of robots is something we’re going to think of as being more versatile than machines.
Nordrum: Exactly. So, you can imagine maybe a general-purpose robot that can be trained to do many different tasks — that’s really the dream of a lot of roboticists — and part of what’s required to get there is borrowing some of the techniques used to make large language models and then tailoring them to robotics.
Robots need a lot of information and different kinds of data on how to move safely in their world to get something useful done. For that, you need not just written texts and images but other kinds of data — maybe spatial data, or data from sensors. You can use all of that as training data in some of these new models that are being developed to train a robot to do more than one very specific task.
Rath: On the medical front, what breakthrough do you find most exciting?
Nordrum: Well, there’s a few that we identified in the medical field. One that has really been a long time coming and that our coverage — you know, we’ve covered this topic for decades now — we’ve named it “stem cell therapies that work.”
We’re talking here about a couple of new treatments for Type 1 diabetes and epilepsy, specifically, that are based on stem cells that are starting to show real promise and which could, if they do work out and pass through clinical trials and such, really help millions of people either possibly get off insulin for good in the long run or specifically reduce the number of seizures they have.
Rath: You said that there were several — talk about some of the other ones in medicine.
Nordrum: Sure. We have another one that we’ve deemed “long-acting HIV prevention medicines.” There are really good HIV prevention medicines out there already, known as PrEP. These can be pills or injections that can prevent you from getting HIV if you’re having sex with somebody who’s HIV positive.
Those have been around for more than a decade, but you do need to take them every day or ahead of your exposure to a virus, and you may not always know when you’re exposed. And, a lot of people, you know, might not remember to take it every day.
Now, there’s a new treatment in development by Gilead that’s had some really promising clinical trial results in the last six months, and it can be administered as an injection once just every six months. It’s showing very, very high prevention of HIV as a result.
Rath: Climate change is obviously an urgent issue. Do we have any technology on this year’s list to give us some hope in terms of breakthroughs there?
Nordrum: We do. We have three — and this is an area where we really think a lot about scale, obviously, because climate change is such a massive problem, and there are so many big industries and very popular products that we all rely on that will need to be transformed or kind of fundamentally rethought. You know, we need to get them to a lower emissions space in order to really mitigate the impacts of climate change.
So, we have three here that, again, the scale really is going to be a big, big part of the story here in terms of why we pick them, but also what they’ll need to ultimately achieve to deliver on this promise of a low-emissions future.
The first one we picked is green steel. The world produces a ton of steel to build all kinds of different things, including in buildings, but it’s a really polluting process with very high emissions. Now, though, there are a couple of companies working on a low-emissions or no-emissions process to produce steel. There’s one plant being built now in Sweden that will begin operations next year that will produce steel using this very low-emissions process.
Rath: You’re talking about scale — changing how steel is manufactured is pretty massive.
Nordrum: Yeah, absolutely. Steel-making, I mean — around 8% of global greenhouse gas emissions right now just come from making steel. It’s a very dirty process. For every ton of steel that you make, you’re producing about two tons of carbon dioxide.
So, you know, it’ll take time. This first plant is going to be really small compared to the total steel making industry as a whole globally. But if we can move in this direction and eventually wipe out that 8% of greenhouse gas emissions that come from steelmaking, that would be a huge change.
Rath: I love this. Is there any way you could give us sort of a lightning round version of the last two?
Nordrum: Yeah! So, just quickly — the last two in the climate category. There’s a trend toward cleaner jet fuel that we’re excited to see because aviation has been a really tricky field to address as well for emissions. Now, in addition to biofuels, which have been around for a while, there are some new ways of making alternative jet fuel that are just starting to scale up production, and also some new policies coming into play that are going to require their use — particularly in the EU, which we think will help drive adoption.
And then, last but not least, we also have cattle burping pills. These address a big problem in agriculture, which is that cows produce a lot of methane. They burp it up as a byproduct of digestion, but they don’t really need to produce it. In fact, they spend a lot of their own energy producing it. Now, there are a couple of companies developing supplements that can be mixed into cows’ food or water that can prevent the cow from burping up that methane and thereby wipe out a bunch of emissions generated from livestock.
