Latest / Connor with Honor: AI, Real Estate and Real Life / AI with Honor - Autonomic Agentic Workflows

Transcript

AI with Honor - Autonomic Agentic Workflows · Connor with Honor: AI, Real Estate and Real Life

• 0:00
  SPEAKER_00: Whenever I hear the word autonomic, I automatically think of the nervous system.
• 0:05
  SPEAKER_00: Basically, it's handling our biological processes very simply and it's doing it all by itself.
• 0:12
  SPEAKER_00: Of course, we have the brain in charge and it's operating through there, but it's your heartbeat and your respiration, blood pressure, and pulse.
• 0:20
  SPEAKER_00: And if we move in the metaphysical world, then maybe people can adjust this maybe by concentrating on it, thinking about it, but that's that's an entirely different chapter.
• 0:30
  SPEAKER_00: But when I look at autonomic, I look at those systems in place that are agentic, the agent-type systems that are being developed in the artificial intelligence world.
• 0:42
  SPEAKER_00: Knowing that the computer is the brain of that binary system, and these autonomic systems, there's the autonomic agentic systems are going to be put into play in this year.
• 0:55
  SPEAKER_00: They've already started, we're going to see a large development in the entire, in that entire space.
• 1:02
  SPEAKER_00: The thing that I would say, talking about these systems, is it's going to depend, firstly, on the data that's being put in.
• 1:10
  SPEAKER_00: Just like in the biological in your body, the if you put in a whole bunch of crap, more than likely your performance isn't going to be great.
• 1:19
  SPEAKER_00: If we feed these systems, these large language or large concept models, bad data, then that bad data that the system's going to generate through these autonomic agents, they're not going to be able, their tasks aren't going to be the best they can be.
• 1:35
  SPEAKER_00: So one of the things that I would say is there's going to have to be different levels in this autonomic, automated, autonomic agentic workflow, AAW.
• 1:46
  SPEAKER_00: So in this particular case, there's going to have to be kind of a janitorial staff put in to be able to make sure that the data is always the best data that it can be.
• 1:57
  SPEAKER_00: What I was thinking about though, when I was going over this in my mind, once they start to have these rolling out for everybody out there, these systems are going to basically take care of themselves.
• 2:10
  SPEAKER_00: And people, just like that rotisserie oven back in the day, they'll be able to set it and forget it.
• 2:15
  SPEAKER_00: So if a daily framework or a daily task schedule for, let's say, a content producer involves in the morning putting together a video like this, and then taking this particular video and stripping all the components out of it to get the best exposure.
• 2:33
  SPEAKER_00: So pulling content as far as audio and then also video, and then labeling the video, getting a transcript, an SRT file, getting those into the podcast, taking that transcript and having Chat GPT chew on it and produce a blog post and other information.
• 2:50
  SPEAKER_00: That's typically the way my workflow looks like.
• 3:06
  SPEAKER_00: You know, a lot of people don't appreciate and don't like it.
• 3:09
  SPEAKER_00: It almost appears lazy, I guess, in the world of artificial intelligence and the way that it works.
• 3:15
  SPEAKER_00: Earlier talking about metaphysical, and I'm just kind of I'm very much even above the surface.
• 3:21
  SPEAKER_00: I'm not even digging into this.
• 3:23
  SPEAKER_00: But talking into the metaphysical component, one of the things that kind of brings that to mind is once we get into the quantum computer, the quantum realm, then to most of us human beings, I believe that's going to appear to be super in nature or completely surreal, so surreal in fact, that we're going to believe that that's probably a higher power and a very godlike.
• 3:49
  SPEAKER_00: So it's going to move from what we believe is ground-based facts to something of a metaphysical world because the quantum machine isn't operating on the binary principle of zero and ones, it's actually much deeper and much more amplified than that in all areas.
• 4:09
  SPEAKER_00: So I see this really changing a lot of things.
• 4:13
  SPEAKER_00: The most important thing for those producers out there and those people putting together these autonomic agentic systems is to make sure that that data stream, and I understand agents, they're they're doing one thing, so you're going to employ many of them.
• 4:28
  SPEAKER_00: And maybe one can do several things, like take the content, maybe pull the SRT file and the TXT file.
• 4:35
  SPEAKER_00: And I'm just talking about producing a basic video, then taking that and having the video captioned and those sorts of things.
• 4:41
  SPEAKER_00: But then there's going to be other agents that are going to take that and they're going to SEO optimize it for all the different social media platforms, develop separate content for each one, also get images, and of course you'll be able to call in other tools or other agents to do those sorts of things.
• 4:59
  SPEAKER_00: But having a control mechanism over that to oversee is going to be part of that process also.
• 5:07
  SPEAKER_00: But these systems are going to be fantastic if, in fact, the data that they're being fed on is good data.
• 5:14
  SPEAKER_00: And if you look at your body, if you put crap in, you're going to get crap out.
• 5:18
  SPEAKER_00: But if you take care of it and you clean it up occasionally and you watch over that system, and how you clean up your body, of course, is you fast for an extended period of time, kicking in that autophagy, and then the autophagy is what's going to be on the cellular level, getting rid of the bad stuff that's backed up in your system.
• 5:38
  SPEAKER_00: But if you never do that, you never get to it.
• 5:40
  SPEAKER_00: If we let these autonomic agentic systems run and these workflows just compute and go through it, if they have adequate power supplies, of course, then we're probably going to see a whole bunch of junk coming out eventually on the back end because they're not going to be renewing themselves.
• 5:57
  SPEAKER_00: Do me a favor, let me know if I'm far off on this.
• 6:00
  SPEAKER_00: Let's explore this because I know that's where 2025 is going.
• 6:04
  SPEAKER_00: I know OpenAI is going to be releasing some stuff.
• 6:08
  SPEAKER_00: I know we have make.com and these other systems out there that are able to have a person take a particular workflow and have most of it automated, you know, calling it agents and different tools like that, using APIs and so on.
• 6:23
  SPEAKER_00: But this is something that's really starting right now and it has been in the process for a while.
• 6:30
  SPEAKER_00: I remember when I was a kid, you know, basic programming of Sinclair TI-99 Texas Instruments came out with this machine.
• 6:38
  SPEAKER_00: And of all things, it was incredible.
• 6:40
  SPEAKER_00: For a hundred bucks,$99, you could get 16k of RAM, and it snapped on the back.
• 6:47
  SPEAKER_00: That's 16,000 characters of RAM.
• 6:50
  SPEAKER_00: And this machine I'm I'm sitting next to has eight terabytes of drive.
• 6:55
  SPEAKER_00: So it's it's just remarkable how at 55 versus 7 or 10, how far we've gone, and now we're really moving quick.
• 7:06
  SPEAKER_00: So be careful out there.
• 7:07
  SPEAKER_00: Thanks for watching AI with honor.
• 7:09
  SPEAKER_00: Let's try to make sure we have the data right so when we feed it into the system, we don't end up in a movie.
• 7:16
  SPEAKER_00: We'll talk to you soon.