Latest / Connecting the Dots / Connecting the Dots - 3-26-2025

Transcript

Connecting the Dots - 3-26-2025 · Connecting the Dots

• 0:00
  Today's tech news moves fast, but understanding
• 0:03
  it shouldn't be hard. That's why Snarful Solutions
• 0:06
  Group brings you Connecting the Dots, where we
• 0:09
  break down what's happening in tech and why it
• 0:11
  matters to you. Now, let's dive in with your
• 0:14
  hosts. All right, so it's March 26th. And let's
• 0:19
  jump into some of the big tech stories of the
• 0:21
  day. You know, you guys sent over some really
• 0:23
  fascinating stuff. Yeah, for sure. And our mission
• 0:25
  here at Snarfle is always, you know, to break
• 0:27
  it down. Like, what does this actually mean for
• 0:29
  you? So we're seeing a lot of action in AI kind
• 0:33
  of globally. Some interesting developments with
• 0:36
  how you might be buying your next phone. Yeah.
• 0:38
  Even some, you know. rumblings in the world of
• 0:41
  crypto so yeah absolutely should we just jump
• 0:43
  right in do it okay cool so let's start with
• 0:46
  the AI scene. This Bloomberg article caught my
• 0:50
  eye. It's about SenseTime, this big AI startup
• 0:53
  in China. They just dropped their financial results
• 0:56
  for 2024. And at first glance, the numbers look
• 0:59
  pretty good. They pulled in about $524 million
• 1:02
  in revenue. That's an 11 % jump from the year
• 1:06
  before. So things are moving. Yeah, that's solid
• 1:09
  growth for sure in any industry, especially in
• 1:11
  a fast -paced field like AI. Exactly. But if
• 1:15
  you look a little closer, the article points
• 1:17
  out that they didn't quite hit the mark that
• 1:19
  analysts were expecting. Yeah, the projections
• 1:21
  were closer to 620 million. So they fell a bit
• 1:24
  short. Interesting. So they're growing. but maybe
• 1:27
  not as explosively as some were anticipating.
• 1:30
  Right, and it seems like they're spending a lot
• 1:31
  too. Their net loss was around $592 million,
• 1:37
  which is actually higher than what was expected.
• 1:40
  So more than they brought in. Yeah, more than
• 1:42
  the estimated $551 million. So they're making
• 1:44
  money, but the costs of developing AI and competing
• 1:49
  in this global race are clearly enormous. Yeah,
• 1:52
  and Bloomberg really highlights that the AI race
• 1:56
  is heating up you know it's not just about building
• 1:58
  cool tech anymore it's a full -on competition
• 2:00
  for global leadership right and this is where
• 2:03
  it gets interesting for you you know you might
• 2:04
  be thinking okay so some AI company in China
• 2:06
  is dealing with big numbers what does that have
• 2:09
  to do with me but here's the thing this competition
• 2:11
  is driving a ton of investment in AI research
• 2:13
  and development absolutely I mean even since
• 2:15
  time which is already pulling in a half a billion
• 2:17
  dollars in revenue is still facing massive costs
• 2:20
  to stay ahead of the curve that's a good point
• 2:22
  It shows how much these companies are pouring
• 2:24
  into innovation. And that's ultimately good news
• 2:26
  for everyone, because it means more powerful
• 2:29
  AI tools down the line. Exactly. Think about
• 2:31
  it. Better voice assistants, more personalized
• 2:34
  recommendations, even breakthroughs in health
• 2:37
  care and education, all powered by AI. It's really
• 2:41
  exciting to think about the possibilities. Yeah,
• 2:43
  for sure. Alright, so let's switch gears geographically
• 2:46
  for a minute. There's this interesting report
• 2:49
  from Reuters about Apple and Indonesia. Apparently
• 2:52
  the iPhone 16 series is hitting the shelves there
• 2:55
  on April 11th. Nice. Which on the surface seems
• 2:58
  pretty straightforward, but there's a bit more
• 3:00
  to the story. Oh, is there? Yeah, this launch
• 3:02
  is actually happening after Jakarta lifted a
• 3:05
  sales ban on Apple products. A sales ban? Wow,
• 3:08
  that's pretty significant. What was behind that?
• 3:10
  Well, from what I gathered, it seems like the
• 3:12
  Indonesian government wanted Apple to make some
• 3:14
  serious commitments to their economy before they
• 3:16
  could sell their products there. Makes sense.
• 3:19
  Yeah. So Apple agreed to invest over $300 million
• 3:22
  in Indonesia, and that paved the way for the
• 3:25
  ban to be lifted. Interesting. So these investments
• 3:28
  could include things like building manufacturing
• 3:30
  facilities, research and development centers,
• 3:32
  or even contributing to the local tech scene.
• 3:35
  Exactly. It's basically a way for the government
• 3:37
  to ensure that companies that profit from their
• 3:39
  market are also giving back and supporting their
• 3:41
  growth. Right. It's a smart strategy to ensure
• 3:44
  that the benefits are shared. Yeah. And for you,
• 3:47
  even if you're not planning a trip to Indonesia
• 3:49
  anytime soon, this story highlights how government
• 3:51
  policies and investments can have a real impact
• 3:54
  on what tech is available where. It's a good
• 3:56
  reminder that we don't always have access to
• 3:59
  the latest gadgets just because a company decides
• 4:01
  to sell them there. Sometimes it's more complicated
• 4:04
  than that. Right. It's about global economics
• 4:06
  and these kind of behind the scenes agreements.
• 4:08
  And it shows how companies like Apple are strategically
• 4:10
  expanding into new markets. all over the world.
• 4:13
  For sure. OK, so let's dive a little deeper into
• 4:16
  AI. Now, I was reading this blog post by Simon
• 4:18
  Willison. He's got a really interesting take
• 4:20
  on Google's new AI model, Gemini 2 .5 Pro. Yeah,
• 4:26
  I saw that one too. He seems pretty impressed.
• 4:28
  He calls it a very strong model. And some of
• 4:31
  the specs are pretty mind blowing, like it has
• 4:32
  a million input tokens and 64 ,000 output tokens.
• 4:36
  Wow, those are a huge number. Yeah. So for those
• 4:38
  of us who aren't AI experts, What do those numbers
• 4:42
  actually mean? Like how does that translate into
• 4:45
  what this thing can do? Well, think of tokens
• 4:48
  as the basic units of information that an AI
• 4:50
  processes, kind of like words or parts of words.
• 4:53
  So a million input tokens means that Gemini can
• 4:56
  take in and understand an enormous amount of
• 4:59
  data at once. OK. Like imagine feeling it in
• 5:02
  an entire novel or a massive research paper,
• 5:04
  and it can process all of that simultaneously.
• 5:07
  Wow. That's pretty incredible. And then the 64
• 5:09
  ,000 output tokens that refers to how long and
• 5:12
  detailed its responses can be generate really
• 5:15
  comprehensive answers based on all that information
• 5:17
  that's processed. So it's not just giving you
• 5:19
  like a one -sentence answer. It's really understanding
• 5:21
  the context and giving you a full, in -depth
• 5:24
  response. Exactly. It's a major leap in terms
• 5:26
  of how much information these models can handle
• 5:28
  and how sophisticated their responses can be.
• 5:30
  And it sounds like its knowledge base is pretty
• 5:32
  up to date, too. Yeah. The reviewer mentioned
• 5:34
  that its knowledge cutoff is January 2025. So
• 5:37
  it's got a pretty recent understanding of the
• 5:39
  world. That's impressive. So it's not working
• 5:41
  with outdated information. It's got pretty fresh
• 5:43
  data to work with. Right. And one of the things
• 5:45
  that really stood out to me was its coding abilities.
• 5:48
  Apparently, they're super advanced. Yeah, I read
• 5:51
  that too. Google is even calling it a thinking
• 5:54
  model. Which suggests that it's not just following
• 5:56
  instructions. It can actually reason through
• 5:59
  problems and come up with creative solutions,
• 6:02
  especially in areas like software development.
• 6:04
  So what do you think all this means for the average
• 6:06
  person? How might this kind of AI power affect
• 6:10
  our daily lives? Well, as these models get more
• 6:14
  and more sophisticated, They're going to power
• 6:16
  the AI tools we use every day. So think search
• 6:18
  engines that understand your questions, much
• 6:20
  better writing assistants that can help you with
• 6:22
  really complex writing tasks, and even coding
• 6:25
  tools that offer more intelligent suggestions
• 6:27
  and automate more of the development process.
• 6:30
  So basically, AI that's not just helpful, but
• 6:32
  actually insightful. Exactly. AI that can really
• 6:35
  understand the nuances of what you're asking
• 6:37
  and give you answers that are truly relevant
• 6:41
  and useful. Okay, so let's shift gears again
• 6:44
  and talk about money, specifically the world
• 6:47
  of digital currencies. Okay. So the Financial
• 6:49
  Times has this report about Fidelity Investments.
• 6:54
  player in traditional finance. And they're apparently
• 6:56
  in the final stages of testing their own stable
• 6:59
  coin. Interesting. So they're dipping their toes
• 7:01
  into the crypto waters. Yeah. And it's a pretty
• 7:04
  big deal. Fidelity is a massive financial institution.
• 7:07
  So their move into stable coins suggests that
• 7:10
  they're taking digital currencies seriously.
• 7:12
  Right. It signals a growing acceptance of crypto
• 7:15
  within the traditional financial world. And the
• 7:17
  article also mentions that Donald Trump has pledged
• 7:19
  to back USD stable coins. So it seems like there's
• 7:22
  some political support for this as well. Yeah,
• 7:24
  that could definitely influence the regulatory
• 7:26
  landscape for stable coins. More political backing
• 7:29
  could lead to clearer regulations and make it
• 7:32
  easier for them to be adopted more widely. Now,
• 7:34
  I know there are probably some people listening
• 7:36
  who are like, wait a minute, what's a stable
• 7:37
  coin? So for those who aren't deep into the crypto
• 7:40
  world, can you break that down? Sure. So a stable
• 7:42
  coin is basically a type of cryptocurrency that's
• 7:45
  designed to maintain a stable value. Unlike something
• 7:47
  like Bitcoin, which can fluctuate wildly, a stablecoin
• 7:51
  is usually pegged to a traditional currency like
• 7:53
  the US dollar. So the idea is that it's more
• 7:56
  stable and predictable, which makes it potentially
• 7:58
  more useful for everyday transactions. Exactly.
• 8:01
  It's kind of like having digital cash that you
• 8:03
  can use for online purchases, money transfers,
• 8:06
  things like that. So if a company like Fidelity
• 8:08
  gets into this and there's political support
• 8:10
  as well, how could that affect the average person?
• 8:14
  Well, it suggests that digital currencies could
• 8:16
  become more integrated into our everyday lives.
• 8:18
  You might start seeing them used more commonly
• 8:20
  for online shopping or even paying bills. It
• 8:23
  could change. how we think about and use money.
• 8:25
  OK, so one last story for today. Let's talk about
• 8:28
  what Microsoft is doing with AI. Sounds good.
• 8:31
  So The Verge has this article about Microsoft
• 8:34
  unveiling new AI deep reasoning agents for their
• 8:38
  365 co -pilot. Trying to make our virtual assistants
• 8:41
  even smarter. Exactly. So they're introducing
• 8:44
  two new agents, one called Researcher, which
• 8:47
  uses OpenAI's deep research model, and the other
• 8:50
  called Analyst, which is based on Microsoft's
• 8:53
  own O3 mini model. And that one's supposed to
• 8:55
  launch in April. Cool. So what exactly does deep
• 8:59
  reasoning mean in this context? Well, it basically
• 9:01
  means that these AI agents can go beyond simple
• 9:05
  tasks like, you know, summarizing a document
• 9:08
  or scheduling a meeting. They can handle much
• 9:10
  more complex requests that require multiple steps
• 9:13
  of reasoning. So they're more like human assistants
• 9:15
  in a way they can think through a problem, analyze
• 9:17
  information from different sources and draw conclusions.
• 9:20
  Exactly. And the interesting thing is that this
• 9:22
  announcement from Microsoft comes right after
• 9:24
  similar AI announcements from Google and OpenAI.
• 9:27
  Right. It seems like everyone is racing to develop
• 9:29
  these more advanced AI capabilities. It's like
• 9:31
  an AI arms race? Kind of, yeah. And the ultimate
• 9:35
  winners are going to be the users because these
• 9:37
  advancements will lead to more powerful and helpful
• 9:39
  tools. So for someone who uses 365 Copilot, how
• 9:43
  could these new agents change their experience?
• 9:46
  Well, imagine being able to ask Copilot to not
• 9:48
  only summarize a report, but also to analyze
• 9:51
  it in relation to other data. You have identify
• 9:54
  key trends, or suggest potential strategies based
• 9:57
  on that analysis. That would be pretty amazing.
• 9:59
  Right. The researcher agent could even do in
• 10:01
  -depth research on a topic and synthesize the
• 10:04
  findings into a concise report, while the analyst
• 10:07
  could help you extract meaningful insights from
• 10:09
  spreadsheets or presentations. Wow. So it's like
• 10:12
  having a team of AI experts at your fingertip.
• 10:15
  Exactly. And it could make us all a lot more
• 10:17
  productive and efficient in our work. Well, there
• 10:18
  you have it, folks. A look at some of the big
• 10:20
  tech stories making headlines today. On the 26th,
• 10:24
  2025, we covered the AI landscape with SenseTime
• 10:27
  Apple's moves in Indonesia, the power of Google's
• 10:30
  new AI model, the potential of stable coins,
• 10:33
  and those smarter AI assistants coming to Microsoft
• 10:36
  365. Copilot, hopefully this deep dive has helped
• 10:40
  connect the dots for you and shown how these
• 10:42
  tech advancements could have real implications
• 10:44
  in your life. We are here to help here at Snarfle,
• 10:47
  so reach out on our website if you have any more
• 10:48
  questions. Absolutely. And we'll catch you next
• 10:50
  time. See you then. That's a wrap on today's
• 10:55
  top stories. Remember, tech doesn't have to be
• 10:58
  complicated, and we're here to help connect the
• 11:00
  dots. If you've got questions or want to learn
• 11:03
  more, visit us at snarfle .com. Thanks for listening,
• 11:07
  and we'll catch you next time.