Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
What humanoid robots need to do to be embraced by humans
71 Plays2 months ago
Mike, Karthik, and Amy are talking about robots! Not just any robots: humanoid robots whose creators are betting will replace human labor in factories, the home, and in services. To do that, the robots will need to overcome human rejection - which might be an insurmountable challenge for the eerie human doppelgangers. Use cases, business models, and the overall weirdness of the human robots are all on the table in this episode.
Recommended
Transcript
Podcast Introduction
00:00:10
Speaker
Welcome everyone to the Innovation Matters podcast where we tackle the big questions about sustainable innovation, like ah would you hug a robot?
00:00:23
Speaker
yeah well We'll see what what our co-hosts have to say here. It's Karthik Subramanian and Amy Hiras from the Lux team on with us. um How's everything going in ah in Amsterdam, Karthik? Good. Looking forward to the fall, still. a Still not there yet. Still not there yet, but in a week. It's going to be there in a week. Very nice. how is How is everything with you, Amy? Everything's great over here, yes. Starting to accept that fall is on its way.
00:00:55
Speaker
I assume that so Anthony is not with us today and I was running late today because it's the first day of school here in the in the Northeast US. I guess schools in Toronto and Amsterdam are probably already started, right? like Like most folks. I think they started yesterday for us. I think last week in Amsterdam. Yeah. we're always The Northeast is always starting late and staying in school late too, but I do like having that extra day or two after the after the holiday weekend before we have to fully dive back into the into the school schedule here.
Humanoid Robots: Capabilities & Tasks
00:01:29
Speaker
yeah but um But Amy, you you had flagged this ah this this story up, so ill maybe I'll let you introduce it and and kick us off and and explain why you and Karthik were talking about hugging robots when I signed on to the that's the call here.
00:01:44
Speaker
Right, yeah, so this is a story that actually Aaliyah O'Brien brought to my attention. She's one of our associate directors. um And this is in the Tech Radar ah story about 1X, which is a Norwegian robotics company. um And just recently in in August of this year, they've unveiled a new model of a humanoid robot, and it's called the neo the Neo Beta home robot. In the unveiling, they have a ah video that shows the the robot in action. The robot kind of looks like, I guess you if you can think of a fencer, um but fence like a fencing uniform, like it's wearing a gray tracksuit with a black mask or like ah a plain black mask for the face, looks kind of eerily like a person.
00:02:30
Speaker
it's It's a little squid game, honestly, though it doesn't have the hood. So it's not not quite the same effect. but yeah Yeah, definitely a bit creepy for sure. Yes, ah creepy. So in the and the unveiling video, they they show a woman who's tying her shoes and the robot is helping her to get ready. So um it picks up her backpack for her. um And before she leaves, they give each other a little bit of a hug. and And so this is why we're talking about whether or not um we would we would feel comfortable hugging a robot. um So this company has been working on humanoid robots for over 10 years now. um They had earlier versions, earlier models, and in those videos, the robots seem a little more jerky in their movements. And so um in the last two years, I think we've this company has paired with OpenAI um and have made improvements. And so this is
00:03:22
Speaker
This is meant to be a ah new and improved model. um And I think there's there's quite a few things we can dig into around ah humanoid robots. um One thing is kind of the state of technology of these robots. What can they actually do? ah What are their limitations? What do these companies expect? um I guess, how do these companies expect robots to be integrated into society if we want to think about it that way?
00:03:48
Speaker
um How much does this tech cost? Is it going to be, you know, um affordable for consumers? And then I think maybe more or one of the more interesting ways to to think about this is how how do consumers feel about these humanoid robots? um Are they going to want to have a robot living in their house? Or I guess we we don't want to call it living. Do we want to do they? Are they going to want to live with a robot?
00:04:12
Speaker
um And I think all of us had an emotional reaction just to watching the video. So we can get into a little bit around the emotions consumers have around humanoid robots, maybe their concerns, their fears, and why this seemed a bit eerie to all of us.
00:04:31
Speaker
Yeah, before ah going in, I just wanted to maybe ask you in terms of just the application. So the video I believe was a very small video where the robot just picks up the backpack and then it's kind of perplexed and trying to you know ask for a hug and then the person comes and hugs it and that's you know kind of it. But is there any indication or information on what are the activities it could do?
00:04:55
Speaker
I think um based on the the article and the previous versions, ah where they had like more extended um ah it was a more extended video that showed different actions. They had it um folding laundry.
00:05:10
Speaker
They had it taking groceries out of a bag and putting it into a cupboard. um So it seems like household tasks, um these kind of washing dishes, ah so like these menial tasks that humans don't um necessarily enjoy doing.
00:05:26
Speaker
um We do already have technology that washes washes dish dishes for us, dish machines and ah you know laundry machines and things like that. There's always a human step involved, but I guess one of the things we can ask is like why why a
Robot Companies and Costs
00:05:43
Speaker
humanoid robot? like Why do we need the robot to look like a human when we do have other forms of technology that don't necessarily replicate a human, but can can do a lot of these tasks for us?
00:05:54
Speaker
Yeah, I mean, at that it's a great question. um And I think it's it is the relevant question for a lot of these these humanoid robotics companies. um And and there's there's a lot of them, right? This is not just um this 1X company. There's ah a lot of, come you know, most famously Boston Dynamics that has been ah developing these kind of humanoid robots and everybody's seen their like funny slash cute slash creepy videos of like robots dancing to pop songs and
00:06:28
Speaker
and stuff like that. um at Companies like Figure AI, Sanctuary AI, Unitree have been bringing out these these types of humanoid robots. um And the costs are coming down. Unitree actually recently came out with one that was only $16,000. Most of these things are, you know, two to, well, are three to five to 10X.
00:06:54
Speaker
that much from from some of these other companies. I was looking at Sanctuary AI. I think their new model is $40,000. So it's a bit more expensive than a dishwasher, too. Yes. Probably not not in the range yet of most consumers. I think it's it's it is reasonable to expect that the costs of this, just as with any type of um electronics, are are going to come down over time.
AI Advancements in Robotics
00:07:19
Speaker
um But the other thing that I think is important as far as the from the technology standpoint and and why we're kind of seeing this this this explosion of of interest in in humanoid robots is because of AI. And ah you know like I said, Boston Dynamics is a good example. They've had these humanoid robots for a long time, 10, 15 years. um But the reason that they're becoming a lot more accessible now is because of the ability to um ah to use AI to help train the robots or make it much, much easier to train the robots. You don't need to program it to do a certain task in ah in a kind of deterministic way. It can basically learn in a more a natural way um you know the to to accomplish certain tasks through kind of repetition.
00:08:07
Speaker
and and imitation. It would, in a way, a bit more analogous to to how a human does. And this is something, you know, even like NVIDIA, right, is the is the big player, of course, in the AI chips. um They are working, they have a whole program around ah robotics,
00:08:26
Speaker
um and and really see this as one of the next big growth areas beyond obviously the large language models that are a big deal today. um And so they're working with Sanctuary and Figure and a lot of these other ah leading players on on developing the the software tools to allow their GPUs and the types of machine learning models that those GPUs are really good at executing.
00:08:57
Speaker
to learn how to ah to to guide robots and and particularly humanoid robots in and doing these various tasks. um So I think that's that's it is an important trend. um And that I think there's reason, you know obviously with all the advances in AI and the the the compute and the chips and and and everything else and the ability to bring down costs, I think there's relatively good reason to be optimistic on the technology front about the the pace of progress that we're going to be able to to see in these areas on the performance and even the cost.
00:09:34
Speaker
of these systems, but the bigger question, and this is where I kick it back to Amy, I think, is is, is anybody going to want this? I mean, particularly for these consumer applications. Yeah, I think this is an interesting question. I guess maybe just one thing about the um the technology thing, this is something I noticed from the video of the the neo ah sorry the the Neo beta, is that in the in the video the the woman didn't actually give a command to the robot so that she was tying her shoe and the robot indicated to the the bag and she said yeah pick it up like or she said yes but didn't actually say pick it up so it was like the robot had anticipated based on the context that the bag would need to be picked up and I think that's where the
00:10:15
Speaker
the AI um for sure um has has really improved the the capabilities of the robots to be able to understand context and what is the likely but is the most likely um action that would be needed in this in this context.
Consumer Concerns and Reactions
00:10:30
Speaker
um Yes, but okay, so so to the question of how do how do consumers feel about this, i've i've I've looked at some of our data um and there's definitely some concerns and I think even just The fact that we we all had like an emotional reaction to watching the video speaks to some of the the the anxiety that consumers feel around humanoid robots. I think um one of the concerns of safety and part of this is not fully
00:11:00
Speaker
um being sure if they can trust the technology to do um the appropriate action. So for example, um and this also came up in in the article, um how does the robot know how hard to hug her, right? So there is the potential that the robot could hug a little too hard and maybe crush the ribs or something like that, right? So um and the the the possibility that robots could do an action that is is um harmful to humans unintentionally, or
00:11:31
Speaker
I mean, I guess we don't want to think about intention here. But um so there's, yeah, the question around safety, um not just in terms of whether or not the robot does something that's unsafe. But could the robots be hacked, right? Is there going to be privacy, security concerns in the sense of if, if you have a robot in your home, and somehow a hacker gets in, and instructs the robot to unlock the doors or something like that, right? So this idea of whether or not the robot could be controlled by someone outside of the home, um whether or not the robot could be used for monitoring purposes or collecting information based on your your your home layout or your your actions within the home. um So there's definitely a concern around whether this might be
00:12:17
Speaker
lead to invasion of privacy, ah monitoring. um And of course, all of this takes place ah kind of within the context of pop culture, where we've, you know, we've all seen the science fiction movies and the books that talk about dystopian futures where robots, um you know, become the overlords or um and consumers are talking, you know, when we look at the the things consumers talk about, science fiction is one of the things that come comes up.
00:12:46
Speaker
um And they're also talking about whether or not robots could become sentient. So again, this goes back to the the the stories, the the science fiction novels and and films where people aren't really sure. There's a bit of a concern of like, what if we take this too far and we lose control of this technology?
00:13:04
Speaker
um And so the other angle, so beyond the safety angle, there is also concern about jobs. This has this has come up like whenever we talk about any um you know paradigm shifting technology, especially with the rise of AI, um concerns that humans or technology will replace, sorry, concerns that technology or robots, humanoid robots could replace humans is is always a concern. so In this context, they're thinking about not having a robot in their home, but whether or not robots might take jobs in factories, whether they might um take jobs in the retail industry, or you know you can imagine a robot assembling your hamburger instead of a human assembling the hamburger. um
00:13:51
Speaker
so even you know I guess it's it's not always can seen as a negative because at the same time there are other dangerous tasks that humans currently do that they think maybe humanoid robots could take over as well. So disabling explosives, dangerous rescue missions, that sort of thing they're they're imagining maybe these could actually um take the place of humans and and and make some professions less less dangerous or reduce the risk in certain professions. So and there's kind of like a ah a good um a give and take there. Any thoughts on this? Yeah, I mean, I think the um the the point about safety ah brings up ah particularly around privacy. It also calls back to, um I think, some of the other products. So some of the things we've been seeing recently that
00:14:43
Speaker
also tend to make me skeptical about these sort of in-home or interpersonal kind of uses of the of the robots. we we've we've um you know Earlier this year, we had the launch of the um the AI pin from a company called Humane, ah which I think we talked about a little bit on the podcast as as well. it's it's a it's It's a pin that you wear on your you know your shirt or your jacket. and and you can press a button to talk to it. And this is meant to be kind of the next device beyond the smartphone. um It was a huge flop, ah did terrible reviews, did very, very poorly in the market. And um a little more recently launched this friend, I think it's just a startup that that did this, ah ah launched a friend necklace where you you can, it's it's a little less ambitious, but you
00:15:40
Speaker
you know you press a button on it and and it and say something to it, and then it texts you stuff. and and But it's kind of it's billed as being your friend, and there's like the the video shows this woman like you know getting feedback on her date about how her date went from this ah from this this AI necklace. um And I think a lot of people have been, I don't know how that one's done commercially, but a lot of the reaction that I've seen to it has been people a little bit creeped out as or or or finding that off-putting. And I think, you know as valuable as robots and and AI,
00:16:17
Speaker
can be for for a lot of tasks. I think that there is, and I'm curious if you see evidence of this in the you know the the cus the consumer conversations that you're looking at, Amy. There's also just this sort of like you know creep like, something that gets to be a little more intimate or a little bit more emotional. People still seem to have a barrier with that to to accepting accepting AI or accepting robots and in that sort of way.
00:16:46
Speaker
Yeah, I think maybe one one other example, and this is ah something that's developed in the last little while, was the the rise of AI is um AI therapists um and consumers have talked about um getting cognitive behavioral therapy from AI chatbots.
00:17:04
Speaker
um um instead of meeting with a therapist in person and there's again kind of the the double-edged sword here for some people this is um it's more affordable than seeing a therapist um so it's and it's more accessible if you live in a rural area or you um find it difficult to find a therapist who speaks your language and that sort of thing. So it gives it gives some some options for people. But at the same time, there's the general sense that it doesn't replace a human and that you know speaking with a human would be top choice. There's there's something about um having a human connection, I think. um Overall, we've seen in the last few years
00:17:44
Speaker
rising concerns around isolation. And I think especially having gone through the COVID pandemic, ah consumers are aware that, you know, these virtual or um um artificial ways of engaging with people don't replace human connection. We are embodied beings, right? So the the physicality being together in person is actually important. And so having like you know this friend pendant with a voice that or a text response to you to kind of mimic a human I think consumers are are aware this is not um actually going to replace human connection or it's a it's a kind of a poor substitute or poor substitute or yeah a lesser substitute, right? It's it's not going to ah replace humans. And so there's something people are really valuing human connection, being able to be together in person, especially having gone through those years where we had to you know communicate primarily through through video or over the phone.
00:18:42
Speaker
um And so using tech to kind of mediate those human relationships has helped people get through and break isolation. But I think people are are really wanting to move back into the the physical realm and connect in person.
Automation's Impact on Lifestyle
00:18:58
Speaker
and And that's one of the things I wanted to ask you, Amy, and and maybe I'm too young to be talking about cathode tube televisions, the the big box ones. So I remember back in the day when I was like a kid and I used to go to my you know grandparent's house, I had to physically go to the TV and press buttons to switch channels. yeah And then then came the the time where you had a remote and then you but you could just lie down on your couch and then just change with the press of a button.
00:19:23
Speaker
And so I feel that as we get more and more automated, there is a sense that we are becoming more sedentary and more lazy. So my question is, are you seeing that come into the picture where there is too much automation in my life?
00:19:40
Speaker
that I'm becoming lazy and I want to get out and I want to experience things. And of course you have the pandemic anxiety the or the pandemic fatigue where you were just too isolated for too long and you just want to change. So are you seeing those ah aspects also being spoken about among consumers?
00:19:57
Speaker
I'm trying to think, I think an this was a project we did quite a while ago and I can't remember who the researcher ah that worked on it was, but there was a discussion of cleaning and consumers were connecting that to mindfulness. So the act, the physical act of cleaning, um although you for a lot of us, that's, it's not that exciting. we We kind of wish a robot would do that for us. For some people, it's it can be enjoyable in the sense that it's a way to zone out and to really just, you you get in the zone, right? And you're washing the dishes and it's actually ah and a break from the the hecticness of your day. um And it's ah some people, you know, it's a physical activity to get your energy out if you've had to sit at your desk all day. So there there are ways consumers are actually finding
00:20:44
Speaker
talking about finding enjoyment in these mundane tasks, especially as there's a movement towards wellness and and thinking about how focusing on the moment um and being present in the moment is is really valuable. Yeah, I mean, that's exactly how I feel about washing the dishes. much It's like I've been working all day, I've been getting the kids back in front of school, I've been making dinner.
00:21:07
Speaker
and and and all this stuff. And then like after dinner, like, hey, I just get to I could put on a podcast, relax, wash the dishes and just, you know, sort of zone out and for ah for for half an hour or so. um But I do think the the area where you would be using these robots where you don't you're not maybe as worried about mindfulness and zoning out or or anything like that is actually in like the factory in in real workplaces. And I think that's that's the place where um There's a lot of opportunity for these these type of systems.
Humanoids in Industry
00:21:42
Speaker
um and i i'm of I'm a bit of two minds when it comes to the to the humanoid robots for that because obviously robotics and factory automation and all that is is is a big is a big theme and ah it been you know it's been something that's been going on for decades. and you know In work that we've done on industrial robotics, the vast majority of that is you know, in these categories like rolling robots, robotic g grippers, robotic arms, cobots, things like this that are, you know, these other different form factors that are sort of more optimized to to a given task. And that's, you know, which makes sense right for any any given task that needs to be done in a factory or
00:22:29
Speaker
or on a job site, like I think construction is an area where there's a lot of a lot of potential for automation ah that that's untapped ah in a warehouse. right If you're going to design the sort of the optimal form factor for for any given task, you would you would not design the human body.
00:22:50
Speaker
um But on the other hand, so much of the ah the way that our infrastructure and and the way the layouts of of of factories and and things today are are built is is designed around the human form. and um And especially if you're going to, as I i think is going to be likely and in the you know at least in the near to medium term, you're going to have both humans and robots in in a lot of these these these factories. you know At some point, you will get to the the lights out factory right where where it you know it's proverbially, it could be run with with no lights on because it's all all robots and automation. There's no humans working inside the factory.
00:23:38
Speaker
um you know I think that that that that is a longer term vision and a lot of in in in most industries. I think with the humanoid form factor, and I was just thinking about, why do you want your human to look like a robot? And when I asked myself that question. For your robot to look like human. Yeah. Oh, yeah. All right. Yeah. That's actually what it's supposed to be. Excuse me for that. But I was just thinking about that. And why should the robot look like a human? And I think the answer is so that
00:24:11
Speaker
We don't get weirded out by a robot that is doing a lot of tasks and it looks more human. So we are comfortable with it. But the more they try to make it human, the more creepy, like the creepier it gets. And so I really don't know how they're going to prevent that negative reinforcement, to be frank.
00:24:27
Speaker
Yeah, I think so. When I was looking at our data as well, I was i was looking at factors that might help allay consumers concerns or what would help people feel more comfortable um with the idea of humanoid robots. One of the things, and I think you already mentioned this, Mike, is cobots. So a robot that doesn't replace humans, but that um requires human collaboration, especially in in things like manufacturing. I think these are already being used. So I think emphasizing the collaboration between ah humans and robots is something that can alleviate some of those concerns around jobs and whether or not like we're going to ah eliminate the need for humans in many industries. And then the other aspect that came up was, and this is this comes to your ah your point here, Karthik, more about the appearance of the robot.
00:25:19
Speaker
um So I think this some of the robots that look creepy um are and the idea of like a robot that is like towering over you, like ah the size of a robot maybe has an impact on whether people would feel comfortable. So maybe there's something about making them smaller. um Making them look cuter might also be um something that helps them look more friendly. Because that this the Neo does not look very friendly, and in my opinion. like so So I think there's there's something about the appearance and maybe the the option for people to customize what their humanoid robot looks like so that they can make it look a way that feels more comfortable to them. But yeah, these are just these are just some of the ideas that came up.
00:26:05
Speaker
yeah And then I think on the on the technology front, I mean, as much as I was relatively optimistic earlier, I think about the pace of progress that we're going to see in this this field of, I think, robotics in general and humanoid robots in in particular.
Challenges in AI Robotics Training
00:26:23
Speaker
over the next several years due to, in large part, the the AI. I think you're going to see less rapid progress ah from from that AI and control and understanding standpoint than you may have seen with with allow with large language models, the LLMs.
00:26:41
Speaker
um for the simple reason that there's not as much not nearly as much data to train on. right The reason that LLMs have been able to advance so quickly and become so ah so capable is because there's, of course, an enormous corpus of written text that's that's available.
00:26:58
Speaker
um You can certainly use you know maybe videos of humans doing certain things to help with the with the trainings of these robots, but I think a lot of the training data needs to be around robots being guided to do a certain task or to, you know like you said, to pick up a bag, you know have it do it 100 different times from 100 different angles and a bunch of different types of bags and things, and then maybe it will be able to generalize from that and pick up.
00:27:25
Speaker
you know any sort of bag and any sort of scenario but it um It does require a lot of that training data and that's I think the challenge that's um that they have and And that's going to be the biggest challenge in the in the near term for these and I think the you know The fact that the hardware is still really expensive there is a bit of a chicken and egg problem there with the training data because the the cost the high cost of the hardware means that you are not having as many of these units in the field so you're not collecting as much of that that training data
00:27:59
Speaker
um But at the same time, the fact that that the training data is limited makes the robots more difficult to use and they're still, you know, easier to program but not ah probably as widely accessible, you know, as something like chat GPT is on the online side, right? So you're not getting ah that sort of volume of user feedback and you're not getting the the economies of scale that you would need to to bring down the cost of the of the hardware.
00:28:27
Speaker
So the I think the data is waiting on the cost of the hardware a bit. And the cost of the hardware is probably also waiting on the cost well on the quality of the models, and and thus on the which which which comes down to, in in large part, the the quantity of the data. um So I think it'll it's it's it's not something that's you know kind of going to be having as as big of an impact like as LLMs are right now in a lot of areas.
00:28:56
Speaker
in the near term. um but you know you are But there is going to be progress. You are seeing like Sanctuary AI, one of the companies I mentioned earlier, I was looking at a case study on um them that one of our our our analysts, Tom Skituke, had done and they were um you know They have robots that are being, ah these humanoid robots, their Phoenix model is being used by Magna, the auto parts manufacturer, and other um industrial companies for for manufacturing you know and and you know being tried out and piloted in the factory. So there's things are moving forward. but um And I think it's it's for for clients and and anyone in industrial manufacturing.
00:29:38
Speaker
ah type of businesses or any types of businesses you know where um you you have those jobs that are that are dangerous or dirty and or or tedious um you know is is worth looking at, but but being a little cautious in terms of how how quickly those are going to come along because of because the data challenges.
00:30:01
Speaker
Yeah, hopefully we are not in the matrix. so We've got some we've got some time before we we ah we get to that point of the Terminator. That's more my my generation of movie reference for that.
00:30:16
Speaker
But um I guess then any of these early adopters of these humanoid robotpe but robots would also need to be okay with the data being collected that or or knowing that their their use of the robot is is being used to train other models. or That's a great question. And and i'm I'm sure it's it's a point of discussion in these ah you know and in in the the deals that these robotics companies are making with the manufacturers. I think from the the perspective of a company like Magna, right theyre they are hoping to collect a lot of data that helps them to optimize the robots for for their tasks and accelerate and their their own progress. um How much of that um
00:31:04
Speaker
you know data generalizes. i I imagine that the you know sanctuary in this case is probably not going to get access to a lot of that data just as you you know like a lot of companies have. Lux has an enterprise you know open AI account, right? And and it's it's set up so that none of our data, none of the things that our analysts ask it gets passed back to open AI to train their models, right? And that's that's pretty common with the LLMs. I think you're going to see mostly similar um i similar arrangements when it comes to to the robotics AI models.
00:31:41
Speaker
but But it is still going to be useful for those those robotics companies just to get some of that real world experience and learning. They may not have access to Magda's data specifically on on those tasks, but I think they'll they will get some valuable learnings about how to train a robot.
00:31:58
Speaker
But even then, for example, with something like Magna that has a specific manufacturing processes or follows a specific sequence of processes, even if it gives out its data, how useful is that data gonna be to train? Because the use case is so specific that you will have to now have a new set of robots that are being trained on a new set of processes for another application. So do you see that kind of data exchange, Mike, to reduce that timeline?
Conclusion and Listener Engagement
00:32:24
Speaker
Or is this not a challenge?
00:32:26
Speaker
i mean I think one of the things we've seen with AI in in a lot of areas is that that I think you see it with LLMs, and I think we've seen it in areas like materials informatics as well, that ah cross-company data sharing is something that that that people are pretty reluctant to to engage in. um And so I think companies will be more focused on you know certainly transferring the learnings you know from one business unit to one application within their company to another. But I think that kind of wider data sharing is is going to be a harder barrier to get over. All right. Well, thanks for the interesting discussion around around that. It's definitely an area we'll be watching. And if if anybody decides to hug a robot, let us know and we'll we'll we'll talk about it on a future future episode of the podcast.
00:33:14
Speaker
um But thanks to everyone for for listening and joining us today. Thanks, Amy and Karthik, for the discussion. If you enjoyed the podcast, please do rate and andub subscribe but and and give us feedback on on your your podcast platform of choice. We always appreciate that. And thanks again for listening and we'll we'll talk to you all again soon.
00:33:40
Speaker
Innovation Matters is a production of Lux Research, the leading sustainable innovation research and advisory firm. You can follow this podcast on Apple Music, Spotify, or wherever you get your podcasts. If you want more, check out www.luxresearchinc dot.com slash blog for all of the latest news, opinions, and articles. so