


10 Jun 2025 · 34m · 11 chapters

Meta’s reported push toward “superintelligence,” plus other tech news briefs and a science segment on CRISPR-edited spider silk.
Dr. Nikki (science explainer; discusses spider genome editing and potential material science applications).
Meta is launching a new AI research lab aimed at systems surpassing human cognition; Zuckerberg is personally recruiting/reorganizing (about 50 hires) and reportedly placing them near him. Meta may partner with Scale AI, potentially investing up to $15B for a 49% stake. Hiring/partnering Alexander Wang (Scale AI founder/CEO) is framed as accelerating “safe superintelligence” efforts, using Scale AI’s safety/alignment evaluation work (including Humanity’s Last Exam benchmark). Discussion questions whether this is “pour money” progress and how “superintelligence/AGI” will be measured.
Scale AI’s safety evaluation; Humanity’s Last Exam benchmark; Apple’s macOS 26 Intel cutoff; Google AI overviews reducing publisher traffic; IBM’s Starling error-corrected quantum computer plan; CRISPR-Cas9 editing of spider silk genes using red fluorescent protein as a marker.
Written by AI. May contain mistakes.
Meta is launching a new AI lab focused on achieving superintelligence.
“Actually, it's not the creepy crawly one.”
Discussion on Apple's potential AI acquisitions and competition with Meta.
“And if you don't understand, Scale AI is the research arm that or has the research arm that does the safety evaluation and alignment lab seal.”
Exploration of how Meta's funding in AI reflects broader industry trends.
“We want to be the company that does it first.”
Discussion on the latest WWDC announcements and the fate of Intel Macs.
“All right, there's plenty more we need to know today.”
Exploring how AI is affecting news publishers' traffic and business models.
“So even though you won't be able to get macOS 27, you will be able to get security updates through 2028.”
IBM's announcement regarding its plans for a large-scale quantum computer.
“And I'm sure the constant changing of algorithms always has something to do with it.”
Overview of Transsion's dominance in the smartphone and EV markets in Africa.
“If not, just know that they're better at error checking.”
Microsoft's decision to move its Build Developer Conference from Seattle.
“Microsoft is relocating its Build Developer Conference away from Seattle for 2026, ending a run of the conference in the company's home city that actually began back in 2017.”
The hosts speculate on the future locations of a tech event and related features.
“The event was also held in Anaheim on Microsoft's campus and in San Francisco.”
A detailed discussion on new research involving genome editing in spiders and its implications.
“Scientists can now edit the genome of spiders.”
Hosts analyze the implications of a recent Apple paper on AI and LLMs.
“I wasn't sure if you had dived into it yet.”

Get the gist, explore an idea or ask about something said.
On Daily Tech News Show, what are the key points of this episode?
Automatic transcript. May contain errors.
0:00This episode is brought to you by Capital One. Capital One's tech team isn't just talking about multi-agentic AI. They already deployed one. It's called Chat Concierge and it's simplifying car shopping. Using self-reflection and layered reasoning with live API checks, it doesn't just help buyers find a car they love. It helps schedule a test drive, get pre-approved for financing, and estimate trade in value. Advanced, intuitive, and deployed. That's how they stack. That's technology at Capital One. Your daily supplement routine doesn't need to be managing a handful of different bottles. Onnit's Total Human brings key research-based vitamins, minerals, herbs, and amino acids together in convenient day and night packs.
0:43One all-in-one product designed to make supplementation easier. The day pack supports cognition, focus, and cellular energy production. At night, the night pack helps you wind down and supports a calm, relaxed state. Less time managing your supplements, more support for your day and night. Shop Total Human now at onnit.com. Grainger knows if you're a procurement manager at an industrial manufacturing facility, you're balancing inventory and budget while sourcing critical supplies across the operation. Too much inventory? Costs climb. Too little? Production slows. Grainger's$2 billion in on-hand inventory helps you get what you need when you need it, so you can help keep budgets balanced and lines moving.
1:26Call 1-800-GRAINGER, click grainger.com, or just stop by. Grainger, for the ones who get it done. Therapy's biggest impact doesn't show up in some dramatic breakthrough. It shows up in the smallest moments. You bite your tongue instead of escalating. You put your phone down instead of firing off that text. You say what you actually mean instead of what would hurt someone else the most. Nobody applauds those moments, but they quietly change your life. That's what growing feels like. And that's exactly what Grow Therapy is here for. Grow is therapy that works, covered by insurance. Sessions average$21 with insurance, and some pay as little as$0, depending on their plan.
2:08There are no subscriptions, no long-term commitments. You just pay per session. Grow connects you with thousands of high-quality, licensed therapists across the U.S., offering both virtual and in-person sessions, nights and weekends. You can search by what matters, like insurance, specialty, identity, or availability. And you can book a therapy session in minutes and begin in as little as two days, not the national 48-day average. See if you could pay as little as$0 for therapy in seconds at growtherapy.com slash ACAST. That's growtherapy.com slash ACAST. growtherapy.com slash ACAST. Availability and coverage vary by state and insurance plan.
2:59this is the daily tech news for tuesday june 10th 2025 we tell you what you need to know follow up in the context of those stories and help each other understand today dr nikki explains why
3:11Dr Niki: we want to edit the genomes of spiders and meta is reportedly going all in on super intelligence Maybe they can team up. Oh, interesting. Oh, God, that's frightening. I'm Jason Howell. I'm Tom Merritt. Let's start with what you need to know with the creepy crawly big story. Actually, it's not the creepy crawly one. That comes later, right? But it's creepy and crawly maybe in a different way. There's something creepy about this. It might make your skin crawly. I don't know. There you go. I think it still works. Meta is launching a new artificial intelligence research lab focused on achieving super intelligence, where it wears a cape and stuff, where AI systems surpass human cognitive abilities.
3:52CEO Mark Zuckerberg is personally leading an aggressive recruitment and reorganization drive at the company, hiring around 50 people for the new team in order to stay competitive in the ongoing AI race, of course. Sources to Bloomberg say that those new hires will actually sit in close proximity to Zuckerberg to keep collaboration efficient and tight. So that's kind of creepy and crawly in itself. Central to the effort is the recruitment of scale AI founder and CEO, the potential recruitment anyways, of CEO Alexander Wang, who would join the new lab as part of the deal that could see Meta invest up to$15 billion for a 49 % stake in scale AI.
4:37That would value the startup up at around$30 billion. So lots of money. Zuckerberg has been frustrated about a few things. Lama 4's performance, delays of Meta's largest model known as Behemoth, just to name a few. And this has really pushed him into what sources are calling founder mode with an increasingly hands-on approach. So super intelligence, it's in the spotlight, I guess, for Meta. Yeah.
5:05Dr Niki: And yesterday at Worldwide Developers Conference, Apple trotted out whatever its strategy is around AI for people to pick at. It is really interesting to see in the wake of that meta making this big announcement because a lot of the rhetoric around Apple is they're behind. What are they going to do to catch up? I have been thinking in my own head. I haven't I don't think I've said this on any of the shows. Apple is just an acquisition away from catching up. They didn't invent the iPod. They acquired Tony Fidel's company to build on the iPod. They didn't invent Apple Silicon. They acquired PA Semi and then used that to develop Apple Silicon.
5:48Dr Niki: So they have acquired a lot of AI companies, but maybe they just need to acquire one more. And here, the day after, is Meta doing that. And not acquiring scale AI, but basically striking a big partnership to take advantage of scale AI. And if you don't understand, Scale AI is the research arm that or has the research arm that does the safety evaluation and alignment lab seal. It is one of the companies behind Humanity's Last Exam, which is a benchmark design to assess advanced systems like superintelligence or AGI. And Alexander Wang was Robert or Robert Sam Altman's roommate during COVID. So there's a lot of tines, a lot of interconnections going on here.
6:37Dr Niki: It is not an insignificant person for Mark Zuckerberg to partner up with in order to supercharge their efforts. Yeah, that's interesting what you mentioned about Apple being an acquisition away because you're right. I mean, all of these big companies, the big, big players, Apple, Meta, Google, all of them, they've got such deep pockets right now. And they are all on a very similar course. They're looking at, you know, where AI is right now. None of them want to be left behind. And in fact, do they not want to be left behind? They also want to be first to whatever the next big thing is. And that's what the super intelligence drive is all about.
7:16That's what this acquisition is all about. It just makes me ask the question once again, which is the goalposts continue to move. What exactly is AGI? What is superintelligence? And how will we actually know if we're seeing it when it actually happens? But of course, they've got the money. They've got the drive. Of course, they're going to put their money where their mouth is and just have a division of the company pursuing that. That's what this is all about.
7:46Dr Niki: Especially because OpenAI has made no secret that that is its goal as well, is to create a safe version of superintelligence. And hiring Alexander Wang, who creates the tools to assess safety, a pretty good way for Zuckerberg to say, yep, that's what we're after too, safe superintelligence. But also, we want to beat you to it. We want to be the company that does it first. Google has DeepMind, you know, and very clearly is after similar things. Demis, you know, on board there, notably. Yeah, and Microsoft, you know, hiring Nobel Prize winners to run its situation there. So, I mean, yeah, you've got a lot of people after this.
8:34Dr Niki: You follow this closer than I do, though, Jason. Like from where you sit, looking at this closely every week, is this just pouring money and hoping for a result or is there some progress? That's what I take from this. This is absolutely, in my opinion, I could be completely wrong, but it really seems like right now, like I said, the goalposts are moving so often. And these companies do have such insane resources that, yeah, it's absolutely a question of let's just throw a bunch more money at this. and see how far it gets us. I mean, apparently Meta is throwing seven to nine figures at its candidates.
9:13So we're talking buckets of money is at the core, you know, driving this because they can, you know, behind that also is the ethos that the bigger model, the better. And so, and what does that take? That takes an insane amount of money. I mean, what it really does is it entrenches these larger players much deeper into the core of the development of these systems and makes it more difficult for the little guys to compete and to catch up. And, you know, maybe there's some considerations there around, you know, a lot of the antitrust movement that we've seen with big tech companies right now. Another thing that comes up for me around this is, you know, we on AI Inside, the podcast that I do with Jeff Jarvis, you know, we chatted with Meta's AI lead, Jan LeCun, a couple of months ago.
10:01So this makes me wonder, like, oh, is Jan not cutting the mustard anymore? Because it really kind of seems like his vision and his leadership is being challenged here. And what does that lead to? But to answer your question, yes, it feels to me like let's just throw a bunch of money at it. Let's make sure everyone knows that this is our stated goal because we, like everyone else, want to be there first, wherever there is.
10:26Dr Niki: The other part of this that I think is interesting is that Scale AI in February agreed to a five-year partnership with the Qatari government to improve government services. In March, it reached a deal with the Department of Defense in the United States to develop the Thunder Forge project. I wouldn't be shocked if Mark Zuckerberg would also like government contracts, not just domestically, of course, but also abroad, whether they're defense contracts or otherwise. There's money there. Absolutely, there is. Yeah, I think you're absolutely right. And so it may be good for investors, good for stockholders, good for buzz to say you're after super intelligence, but possibly under no illusions about how fast that will happen.
11:15Dr Niki: However, it's a good way to get some lucrative contracts. Yeah, indeed. And capitalize on kind of the moment in attention around AI. Because, I mean, the flip side of this is I think there's a lot of people very high up saying, oh, you know, AGI is just around the corner. Super intelligence. It's within grasp. That's why we're doing this. Yet the models themselves that we do have right now are incredibly imperfect, like comically imperfect to a certain degree. Will we get there? probably eventually longer term. I don't think we're anywhere near that. But a story like this certainly capitalizes on the attention around AI.
11:57Dr Niki: And that's part of it too. And it probably achieves multiple things. Another one is as clunky as some of the models are right now, they get better fast. They've been getting better in front of our eyes faster than most technology advances. And Meta doesn't want to get left behind in that. So if you push the outer edge, you're likely to catch up in the other parts too. Indeed. DTNS is made possible by you, the listener. Big thanks to Tony Glass, Philip Less, Howard Yermish, and we got a bunch of brand new patrons supporting us at patreon.com slash DTNS. Hans Solo, Jonathan, StreetbreakerX2, and big YouTube supporter, Laminar Rainbow.
12:37Dr Niki: Thanks to everybody for jumping in on that.
12:45trackers are all in different places. That's where ChatGPT work comes in. Give it the goal and connect to the tools where work happens, like apps and files, and it can help pull the pieces into a draft you can review without you having to re-explain the project at every step. For example, ChatGPT work can gather team feedback from Slack and use it to suggest updates to a project brief in Google Docs. You stay in control reviewing the draft and deciding what moves forward. Put ChatGPT to work on your most ambitious ideas and projects. Get started at chatgpt.com by selecting work mode. Available on Plus and Pro plans.
13:22I used to be the person who hunched over my laptop mid-meeting, typing furiously to catch every word, and missing half the conversation because I was too busy transcribing it. That's the problem Granola fixed for me. Granola is an AI-powered notepad that makes your meeting notes useful. It captures what happens in your meetings and turns it into clean, structured notes with decisions and action items pulled out and made easy to find. And the best part? It integrates seamlessly with how you already work. No setup, no configuration, no friction. Before Granola, my post-meeting routine was triage.
13:5420 minutes after every call decoding my own shorthand, figuring out who owed me what. Now the notes are just done. Organized action items pulled out before I've even closed my laptop. You get better notes in one place and you get to actually stay present instead of typing the whole time. Once you try it on a first meeting, it's hard to go without. You can try it totally free. Head to granola.ai slash dailytech. That's granola.ai slash dailytech to get your time back. Go to granola.ai slash dailytech and try it on your next call. Being a serious business owner takes hard work, patience, and a drive to keep moving forward.
14:30So it's important to find a serious business card that goes the extra mile. With the VentureX business card from Capital One, you earn unlimited double miles on every purchase. And with big purchasing power, you can spend more and earn more. This is your sign to take your business to the next level. The Capital One VentureX Business Card. What's in your wallet? Terms apply. See CapitalOne.com for details. Mod.com is changing the way people battle excessive fatigue. Ever hit a wall in the middle of the day and a third coffee just doesn't do it anymore? Here's what's actually happening. Coffee only briefly blocks adenosine, the chemical that tires you out.
15:10When it stops working, your system is flooded and you crash. Modafinil is a prescription treatment that works in a completely different way. Modafinil works by stimulating multiple receptors in the brain associated with wakefulness, including dopamine, serotonin, norepinephrine, histamine, glutamate, and orexin. It's a more complete solution that gives you 10 to 12 hours of steady focus and energy. To learn more, go to mod.com.
15:39Dr Niki: Fill out your intake form, meet with a licensed provider, and if approved, ModAlert ships straight to your door. Go to mod.com to get started. That's M-O-D dot com. When demand slowed for HydroJug, Amazon Ads rising star Hayden Wadsworth pivoted. When our bank account hit near zero, my brother Jake and I had one last shot to save HydroJug. So we engineered the Traveler, a leak-proof, coupler-friendly Tumblr. Amazon ads enables a business to be able to expand their product shelf because of the visibility and the reach you can get. Each week, our new to brand customers make up about 95 % of all of our sales.
16:18Watch HydroJug's story at advertising.amazon.com slash rising-stars.
16:26All right, there's plenty more we need to know today. Let's get to the briefs.
16:29Dr Niki: Yes, it's WWDC hangover day. Not that WWDC is over, but the keynote is always the big event. And we did talk a lot about the big announcements. There's always a couple of things that trickle out afterwards. For example, Apple announced that macOS 26, a.k.a. Tahoe, will be the last version of macOS to support Intel-based Macs. I know. It seems like only yesterday that we had the M1 chip launch, but only a few Intel Macs are even going to be able to support the Tahoe update. So if you have a 2019 16-inch MacBook Pro or 2020 13-inch MacBook Pro with the four Thunderbolt ports or a 2020 27-inch iMac or a little bit older, the 2019 Mac Pro.
17:15Dr Niki: So basically 2019 being your older limit, you will be able to get Tahoe older than that. You're not going to be able to upgrade to this. However, Apple did say it will commit to three years of security updates for those machines. So even though you won't be able to get macOS 27, you will be able to get security updates through 2028. At least there's that. Wow. Wow. Intel fading. Like, yeah, you're right. It seems like only yesterday, but it was five years ago. But, you know, this is one of those examples where really I see this all the time now where the COVID gap came along. Oh, yeah. And it really has warped my perception of time because there's like two, two and a half, three years there that feel like a completely different planet to me.
18:04Dr Niki: Yeah, no, you're absolutely right. I mean, part of it is just time speeds up the older we get. But I think that put a little extra pressure on the accelerator there for a couple years. What are you saying? I'm old, Tom Merritt? No, no, only talking about myself. You are forever young, Jason. Thank you. Google's AI powered search features, including AI overviews and the recently kind of rolled out AI mode chatbot are significantly reducing traffic to news publishers. This is according to a Wall Street Journal report. The New York Times has seen its organic search traffic drop from 44 % to 36.5 % over the past three years, according to reported data from similar web.
18:46Business Insider has seen its website traffic decline 55%. 55%, wow, that's a lot, in the same time frame. These changes are prompting news outlets to see an urgent need to explore new business models and to continue to forge content sharing deals with AI companies like we've seen recently with like New York Times and Amazon to survive the shift in consumer habits and AI tools. Everything's a changing, whether Google wants to admit it or not, because Google still says, oh, no, they're high quality clicks. It's OK.
19:18Dr Niki: These numbers were declining before chat GPT. That's true. So they may have accelerated since ChatGPT launched, and that may or may not have to do with ChatGPT's usage. Although if it is, it certainly is a feather in OpenAI's cap that there are that many people using ChatGPT or Anthropic or other tools like Perplexity. I think this was a problem that the publishers already had. I'm not 100 % convinced that fighting Google over paying for links isn't also a small contributor to the decline as well, because Google has been pretty intransigent in that. And I'm sure the constant changing of algorithms always has something to do with it.
20:03Dr Niki: It also could be people just doing fewer searches for news these days. I think it's smart for these publishers to look for a direct relationship with their customers. I think that's smart. I think it's smart for them to pursue the licensing deals and try to get their content front and center in the chatbots as usage increases. But I have to say, even as a podcaster, we notice, Jason, that you can't just expect people to find your podcast in a search and go to your website anymore. You have to be available on the platforms. So I imagine that has a lot to do with this as well. In fact, the fragmenting of platforms from X to Mastodon to Blue Sky makes it harder for these companies to create a cohesive strategy.
20:50Yeah, that's absolutely true. There is a lot of fragmentation and a lot of different directional impacts. It's hard to attribute it to one thing. I think the fact that this really covers the last three years is kind of the tell right there. It's like, OK, well, these tools haven't been around in the way that they are right now for the last three years. So there was something brewing prior to that, which is not to say that it isn't part of the impact. It almost certainly is a piece of the impact, but it's probably not the whole pie.
21:21Dr Niki: Yeah. When you see that 55 % decline, I'd be curious to drill down a little more and find out how much of that could you really attribute to the chatbots and how much of it is other factors. Business Insider put up a paywall, and I'm sure that cut down a lot of people going to Business Insider. It certainly did for me. Absolutely. It did for me, too. Yep. IBM announced plans to start building Starling, which the company claims will be the world's first large scale error corrected quantum computer to launch by 2028. If you're like, wait a minute, didn't Google say that they would do the same thing?
21:55Dr Niki: Didn't some other companies say they would do the same thing? Yes, everybody's trying to be this. But IBM is the latest. And all of these efforts are slightly different. Cloud access is expected to become available by 2029. The machine will feature 200 logical qubits capable of performing 100 million consecutive logical operations with high accuracy, which would be a big improvement over current quantum computers that handle a few thousand operations before errors arise. So 200 qubits is pretty good. IBM's solution uses a low-density parity check code, requires 12 physical qubits per logical qubit, along with real-time error detection capabilities.
22:36Dr Niki: If you know what that means, that will mean something to you. If not, just know that they're better at error checking. Beyond Starling, IBM has plans to build another. In fact, they've got a bunch. There's Loon in there. There's a Cockatoo. And then Blue Jay will have 2 ,000 logical qubits and be able to process a billion logical operations. They like the bird names at IBM. A lot of people still skeptical that these will even be commercially useful, but they are certainly progressing. Yeah. You know, again, it's kind of like super intelligence. I guess you got to start somewhere. Yeah. And they'll be good for something, whether that's something is, you know, a thing that an enterprise company will want to pay a lot for cloud access for or whether it's more research oriented remains to be seen.
23:27Yeah. Long enough timeline. I'm sure it does amount to something. But, you know, again, it's that question mark of like, OK, yeah, but how long until that becomes the reality? We really have no clue. You know, again, they just they want to be where everybody else is going to be, too. It's it's very similar to the race right now. Sony announced that its PS5 DualSense controllers will receive a firmware update sometime this year that's going to enable support for multiple simultaneous Bluetooth connections. Users will be able to pair a controller with more than one device at a time. So like PS5, PC, smartphone, tablet, and that new functionality will bring switching seamlessly between those connected devices to the controller.
24:10That way you don't have to like repair every time you want to connect to a new device. No word on exactly when this new feature will roll out though.
24:19Dr Niki: Oh, repairing, not having to repair sounds wonderful to me. And it really shows that we are moving into a world where the Xbox and PS controllers are just becoming the controllers for everything, right? Because you can pair them with stuff and so many things are interoperable with them. That's great. Love it. Shenzhen's Transhen, which is spelled with two S's in case you're wondering, is the leading smartphone maker in Africa. If you've heard of brands like Techno, T-E-C-N-O, Infinix with an X at the end, ITEL, not an N. Those are big brands on the subcontinent. Transcend has about, among those brands, half the smartphone market in African countries.
25:04Dr Niki: And now it's trying to become the continent's dominant electric vehicle maker. Transcend introduced its Tankvolt e-bikes in Uganda in 2023 and is now announcing its expansion into Nigeria, Kenya, Tanzania, and Ethiopia. It's the number three EV maker by sales in the African market and expects to be the market leader by the end of next year. Bikes are among the cheapest you can get at about$1 ,500 U.S. They come with built-in or swappable batteries, and Transyn offers leases for the batteries as a service to encourage people to create third-party swapping stations. In fact, they've got one going in Tanzania with service in Uganda to start by the end of the month.
25:45Dr Niki: The company owns the supply chain. So they manufacture the bikes, they distribute the bikes, like they go from making the bikes to giving them to the consumer and sell directly to consumers, which is a big advantage for Transyn. A lot of fleet customers are buying these. So you're getting a lot of governments, a lot of big companies buying them. Its big challenge will be expanding that charging and battery swap infrastructure. It faces stiff challenges from Benin's Spiro, Irwanda's Ampersand, and Ethiopia's Dodai. Transian. Transian. Yeah. This is a company that I'm not super familiar with. When you see Android dominates the world market, Transian is a big chunk of that because Android phones in African countries are pretty much dominant and Transian makes the majority of them.
26:35Yeah. Yeah. Well, actually, when I was at Mobile World Congress in Barcelona, and I'm sure I passed by the booth and didn't realize it, but I definitely checked out Techno, which was another brand that you just mentioned. That's Transient.
26:46Dr Niki: So if you checked out Techno, you checked out Transient. Okay, well, there you go. Techno, Infinix, and ITEL are the three big brands under Transient. You just never hear about Transient. Now I get it. Yeah. Okay. Yeah, and I mean, I was actually really impressed by some of the stuff they had going on there. The Techno Spark Slim, the super duper slim ceramic phone that they had was beautiful. Yeah. Cool stuff. They make really good phones at a smaller price point, and they focus on Africa. They dominate in Africa. They're available elsewhere in the world. I think India is a big market for them, too.
27:16Dr Niki: But, yeah, something that if you're in Europe and North America, you probably don't run into as often. Indeed. Microsoft is relocating its Build Developer Conference away from Seattle for 2026, ending a run of the conference in the company's home city that actually began back in 2017. all future reservations at the Seattle Convention Center have reportedly been canceled, though Microsoft hasn't yet announced where next year's Build 2026 will or might be held, or if it might instead become an online event, whatever the case may be. Reports claim that security and safety issues related to the event being held in downtown Seattle are at least part of the reason behind this switch.
28:01And you might remember before this was moved to Seattle, the The event was also held in Anaheim on Microsoft's campus and in San Francisco. So who the heck knows where it's going to end up? Maybe just online.
28:13Dr Niki: If I had to bet, it won't be online. Well, it would be online because they're all online, but it won't be only online. I'm going to put my money on Redmond. They're going to just move it right back to the Microsoft campus because they want to keep it in the area. And that way they can impress developers with a little more tour. my second, my number two, you know, my win place and show would be, I don't know, maybe San Francisco. Maybe they'll get enticed back to San Francisco or Anaheim. I can't, I can't tell. It seems like it'd be one of those though. Yeah. Yeah. I agree. Threads is about to get its own direct message solution.
28:50Dr Niki: Finally. Company is testing the new feature beginning this week. According to Mark Zuckerberg, users will finally be able to send DMs without leaving the platform to go to Instagram. If you're like, wait, you can DM on Instagram. Yes, now you won't have to go to Instagram. You can DM right from within threads. New inbox will be accessible through an envelope icon in the app's navigation bar. Hong Kong, Thailand, and Argentina are the first places to get the new feature with more markets to follow. Just one of those updates that you're just like, really? It took this long? DMs seems like a table stakes feature to me.
29:22I'm just kind of surprised by it.
29:23Dr Niki: I mean, everybody hates DMs but loves DMs. Yeah, yeah, totally. are a challenge to manage. So I guess I shouldn't be too surprised that they took. But the other part is like, yeah, but you have a policy. You have a system on Instagram. Yeah. Every product manager sighs when you say, why didn't you get to this earlier? Because there's a list a mile long. They're like, you don't know. You just don't understand. Those are the essentials for today. Let's dive a little deeper into another topic. Scientists can now edit the genome of spiders. Yeah, Dr. Nikki explains the practical reasons why they would even want to do this, but also whether it can make you into Spider-Man.
30:07Dr.
30:07Dr Niki: Nikki, thank you for joining us to talk spiders. Oh, always here to talk about spiders. In this case, specifically spiders from the University of Beirut in Germany. Not Mars, Germany. Dust spider. so uh these researchers are editing spiders for the first time and you may think why because you know we've heard about edited mosquitoes and wasps um but this is a crispr cas9 edited spider and just to refresh you on crispr cas9 if you've been living under a rock uh no i'm just kidding uh crispr cas9 won its inventors the 2020 nobel prize it is a gene editing tool that helps you either cut out or insert or both specific pieces into a genome.
30:54This is going to be called a knock-in or knock-out function of certain genes.
30:58Dr Niki: Yeah, it's whenever you hear gene editing, most of the time they're talking about CRISPR-Cas9. Why would they do spiders? Why spiders next after all of this? So spiders, because they're kind of a difficult challenge. They have an extra tricky and complex genome. They're also not the best lab animals. They're cannibalistic, so you can't house them together. And they're a little bit sensitive to how they're kept. So they're generally maybe a little bit harder to work on than mosquitoes, which is why it took so long. Also, you probably know that spiders have really interesting property, which is they make silk.
31:33And silk has a lot of interesting things that are useful for different types of research. We know that it's five times stronger than steel, for example. And so it would make sense that we want to see if we can edit silk and see what we can get out of it.
31:49Dr Niki: Yeah, yeah, that makes sense. Okay, so we didn't start with spiders because they eat each other. But we finally got around to spiders because dang it, we want that silk. So what did they actually do? What genes did they edit and why? So they edit the specific protein production part of the silk genome. So we're editing the silk protein. They did this by making a solution with this gene editing function, injecting it into oocytes, which are just egg cells that were in unfertilized in a female spider. They waited till the female spider didn't eat the male and reproduce and made genetically modified spiderlings.
32:31Love that word. And the way that they're able to tell whether this worked or not is just that they included a red fluorescent protein into the edit so that if they, you know, turn on a fluorescent light and saw the protein, they could see whether or not it worked. They didn't really change anything structural or crazy about the silk. This was kind of more of a proof of concept to show that we know how to target the specific spider silk protein area and that we can change it. What we want to change it into is now open to everybody.
33:00Dr Niki: Got it. Are we going to be Spider-Men? so as of right now i'm gonna say no unless you want to make spider-man a red fluorescent silk dress uh we might be closer to that you can make spider-man's outfit spider-man's outfit fluorescent we could do it maybe i need a lot of spiders but we could try yeah because a nice red spider-man fluorescent outfit would work uh what what is this pointing to um i think there's a lot of interesting promising functions here for material science so maybe we want to make silk a certain color that's more of like a fashion thing but maybe you can make it stronger maybe you can make the proteins stickier there's a lot of things that i think now that we've figured out the genome and what area to edit we might be seeing an explosion in spider silk research maybe different like properties of clothing like not like a like a kevlar type spider silk outfit i don't know maybe we can increase spider silk production.
34:01So there's a lot of different ways that this can evolve. I'm excited to see where it goes.
34:05Dr Niki: I imagine you could even do some more slightly mundane things like just increase output of silk. Yeah. And I wonder if this maybe depends on the spider's biological capacities. So I'm not sure about that. And you'd probably have to edit more things within the spider. I don't know. But we're able to edit the spiders now. So, you know, the possibility is there. Glow-in-the-dark silk on its own, though. That's kind of cool, honestly. Yeah, I wonder if we're going to see this pop up in the next Spirit Halloween. Yeah, I actually have glow-in-the-dark spiders on fake spider webs that we put up at Halloween.
34:45Dr Niki: You beat them to it. Yeah, just making, well, no, I just need the spider webs to be glowing now and be made of real silk, too. This is, yeah, this is really interesting. And I think, especially with material science, a lot of times it really is you need to make the thing, make the material, so then you can play around with it. And I feel like that's what we're at. We're at the step where like, okay, we can now do this. What should we try with it? Yeah, looking forward to all these new applications. Well, Dr. Nikki, if folks want to find more of what you do, where should they go? Of course, people can follow the thread and find me on the web.
35:21Oh, I'm over at NicoleAckermans.com and the same on Blue Sky.
35:27Dr Niki: Thanks, Nikki. What would you like to hear us talk about on the show? One way to let us know is our subreddit. You can submit stories and vote on them at reddit.com slash r slash daily tech news show.
35:43This episode is brought to you by Capital One. Capital One's tech team isn't just talking about multi-agentic AI. They already deployed one. It's called Chat Concierge and it's simplifying car shopping. Using self-reflection and layered reasoning with live API checks, it doesn't just help buyers find a car they love. It helps schedule a test drive, get pre-approved for financing, and estimate trade in value. Advanced, intuitive, and deployed. That's how they stack. That's technology at Capital One. Some days, getting the work done isn't the problem. Staying locked in is. Alpha Brain is Onnit's daily nootropic formula for cognitive support, designed to support memory, focus, attention, and mental clarity.
36:25Its formula with science-driven ingredients include theanine and huperzine A to support focus and memory, and it's caffeine-free. So when you're deep in the work, Alpha Brain helps support enhanced productivity and tune-out distractions. Get focused on what matters. Shop Alpha Brain now at onnit.com.
36:47And we end every episode of DTNS with some shared wisdom. Today, Mike tips us off to an interesting take on Apple's paper about LLM complexity that we discussed on the show yesterday.
36:58Dr Niki: Yeah. Did you see this paper? I wasn't sure if you had dived into it yet. I haven't dove like super deep into it, but I've read pieces. I've read some analysis. Yeah. And essentially, I think the way we described it yesterday is it's a paper that says there's a certain level of complexity after which LLMs don't improve anymore. They kind of fall off a cliff, but that doesn't mean LLMs are useless. It just means there are other tools that might be better at other things. Mike wrote in and said, I wondered if you had seen this video from Nate Jones regarding the Apple AI paper and his perspectives on what perhaps is being missed in the angst going on regarding the paper.
37:37Dr Niki: Now, Nate Jones is responding to a lot of folks on social media saying, Apple declaring LLMs dead and either taking the position of, see, I told you LLMs suck or Apple's crazy. This is why they're no good at AI. Here is an excerpt from the video description that Nate Jones wrote. He said, I explain why Apple's new paper doesn't kill AI reasoning. It simply shows that small models, stripped of tools and extra inference time, fail on very hard logic puzzles. The study's narrow setup sparked a meme frenzy declaring AI useless, but those hot takes ignore everyday systems that rely on bigger models, Python, and Search.
38:23Dr Niki: The real takeaway is architectural. Production agents must recognize their own limits and escalate to heavier resources when complexity spikes. Rather than eulogizing AI, we should build standardized call for help triggers and rerun these tests with modern tool enabled models. Thank you, Mike, for sending along this YouTube video. It's really a really good watch and a good read, even just in the description there. But I think this jives with my take on this as well, which is this shows LLMs need to know where to call for help when they're agentic. And that's kind of where we're going with stuff.
39:02Dr Niki: So far from saying LLMs are dead, it's like, oh, this is how LLMs will be used in a suite of tools going forward. Yeah, yeah. It's a real challenge for AI models, you know, to know when they get something right, know when they get something wrong, and know when to reach out for help, call for help like that game show. It's the Kenny Rogers problem. They need to know when to hold them. When to hold them and when to fold them. And when to fold them. So, so very true. But yeah, I think it's definitely a challenge that that should be solved. And also kind of, you know, like he pointed out, the study had some pretty serious restrictions in there.
39:41And so if it was limited out of the gate, then, you know, of course, you're going to get a certain level of output out of that. And there are a lot of people who feel very strongly either for or against AI that this was kind of one of those moments where it either confirmed or confirmed their side of it, depending on how you look at it.
40:02Dr Niki: Yesterday on the live stream, Justin Robert Young had nothing nice to say about this study. I think it's suffering from mainstream views on academic papers. A lot of times academic papers are just trying to show a very small thing that's part of a much bigger puzzle. And sometimes they're even saying, we think this is true, but let's study it and make sure it's true. This strikes me as an example of that where they're like, I think small models, if you cut off their inferencing, if you cut off their tool access, have a limit. Let's figure out what that limit is because that'd be handy to know when we're implementing them in other cases.
40:37Yeah, indeed. Well, what are you thinking about? What are you inferring? If you got some insight into a story, share it with us. Feedback at DailyTechNewsShow.com. Thanks to Dr. Nicky. Thanks to Mike for contributing to today's show. And thank you for being along for Daily Tech News Show. And good news, if you want to keep up on music news, Daily Music Headlines is now available on an Amazon Flash Briefing. Go ahead and add it now. We'll see you next time.
41:08the dtns family of podcasts helping each other understand diamond club hopes you have enjoyed this program this episode is brought to you by capital one capital one's tech team isn't just talking about multi-agentic ai they already deployed one it's called chat concierge and and it's simplifying car shopping. Using self-reflection and layered reasoning with live API checks, it doesn't just help buyers find a car they love. It helps schedule a test drive, get pre-approved for financing, and estimate trade-in value. Advanced, intuitive, and deployed. That's how they stack. That's technology at Capital One.
41:52Some days, getting the work done isn't the problem. Staying locked in is. AlphaBrain is on its daily nootropic formula for cognitive support designed to support memory, focus, attention, and mental clarity. Its formula with science-driven ingredients include theanine and huperzine A to support focus and memory, and it's caffeine-free. So when you're deep in the work, AlphaBrain helps support enhanced productivity and tune out distractions. Get focused on what matters. Shop AlphaBrain now at onnit.com.
IBM announced plans to build its major Quantum computer called Starling, and Dr. Niki explains why we want to edit the genomes of spiders.
Starring Jason Howell, Tom Merritt, and Dr. Niki.
Show notes found here.
Hosted on Acast. See acast.com/privacy for more information.