0:38sarah: Thank you so much, Alex. It’s a real pleasure, as you know. I’ve been working in the space for some time, but also excited about Prisado’s role in this space. So thank you again for having me. My name is Sarah Luger.
0:53I got my Phd. At University of Edinburgh many years ago in artificial intelligence, and I would be.
1:00sarah: I’m being quite honest to say that I’m surprised and excited about the developments in this space. In the past I’ve worked at startups. I’ve worked at Ibm building a precursor to the Ibm. Watson jeopardy challenge robot.
1:18sarah: I’ve also been at Orange Silicon Valley for 5 years. We’ve worked on numerous topics, including voice by metrics. Chat bots call center technology.
1:28and of course.
3:04sarah: okay?
3:06sarah: So I think
3:09sarah: for the the average person out there. General, today I
3:15sarah: is AI, where the output resembles human content. It resembles a language that is either
3:25that seems like it’s constructed by a human
3:29sarah: or technically generative AI systems are based on algorithms that learn from the a vast amount of input data. And the most recent cases that we’ll dig into. That would be all of the digital data that’s on the web as well as some knowledge bases knowledge bases being things like Wikipedia that give structure and associate terms. And
3:54sarah: some apparent meaning to this, this vast sea of of language data. And so
4:02sarah: what’s going on under the hood is that there is
4:05sarah: this vast amount of data is being used to learn the patterns of how we as humans speak
4:14sarah: and how we write, and with innovations, both from Google’s 217 transformer paper incredible compute innovations
4:26as well as just ongoing neural networks developments.
4:33sarah: There’s the possibility, as many of us have have now tried since November thirtieth, 2,022, when Chat Gpt was launched
4:41sarah: to engage with an generative AI system in a way that most people had not engaged with an AI system, you know. Perhaps in the past you had AI, a secondary characters in a video game, you know, or there’d maybe been some predictive analytics in an enterprise app
5:02sarah: application you were using. But the core of generative AI is using these patterns of words at at a vast scale. That then for us makes it seem like
5:15sarah: this computer, is
5:17sarah: it? Almost? It’s almost a a human like content that’s being output. And it’s really a powerful difference between
5:25sarah: systems from even gosh, even 6 months ago, right? We’ve had 7 months ago. We’ve had a seat change. And your second question is, who is it
5:37sarah: most?
5:39sarah: Who is it most important? Who is? Who is it most important? For
5:44sarah: who is it most important? For well, right now the
5:49sarah: we’re in. We’re in the the hype cycle, and it’s and it’s a little bit of it’s important for everyone.
5:57This is great for everything.
5:59sarah: And I respect the hype as someone who’s in Silicon Valley, because I understand the role that it plays and the duality of of
6:11sarah: of how we get an investment and how we build prop products and how we have to compete with other hype cycles, be them metaverse. most recently in blockchain.
6:22sarah: But I think that this is really important for
6:26sarah: creating customer-centric tools
6:46sarah: human seeming
6:49sarah: engagements that can be created for their customers. So
6:54those are the 2 areas that the people I see that it will most of affect. But then I want to also flip that and say.
7:02sarah: I think the Holy Grail of enterprise.
7:06sarah: Innovation that isn’t as shiny and sparkly as some of the other. you know. Key terms I’ve just mentioned
7:20sarah: so the ability to search through a company’s resources, to answer questions for employees, or answer questions for employees that are then passed on to customers. I think that that is
7:36sarah: really key cause it will help you and I do our jobs better and reduce mundane tasks and reducing
7:45Monday and task.
7:48sarah: It’s something that AI in general is, you know, aims to do any technology. We try to elevate our
7:55sarah: our work tasks up
7:58the the difficulty chain. So we want, as humans to not do the same thing every day, but to understand patterns
8:10sarah: and reduce repetition
8:14sarah: and do more and more challenging tasks, and those more challenging tasks are very hard for computers, so don’t fret. Many of us will still have jobs. On the other hand, those lower level tasks are.
8:30sarah: really great opportunities for computers to come in generative AI systems to come in and support us.
